Merge pull request #72 from modelstudioai/chore/list-voices

feat(omni,speech): add --list-voices and fix cosyvoice voice ID
chore(release): 1.4.2
2026-09-14 19:49:23 +08:00 · 2026-06-24 10:40:10 +08:00 · 2026-06-24 10:35:26 +08:00 · 2026-06-24 10:29:15 +08:00 · 2026-06-24 10:16:20 +08:00 · 2026-06-23 17:08:37 +08:00
139 changed files with 7462 additions and 1342 deletions
@@ -0,0 +1,185 @@
+name: Bug Report
+description: Report a bug in bailian-cli (bl)
+title: "[bug]: "
+labels:
+  - bug
+body:
+  - type: markdown
+    attributes:
+      value: |
+        Thanks for taking the time to report a bug.
+
+        **Before submitting:** search [open issues](https://github.com/modelstudioai/cli/issues?q=is%3Aissue+is%3Aopen) for duplicates.
+
+        **Security:** redact API keys (`sk-...`), console tokens, internal URLs, and business prompts before pasting output.
+
+  - type: markdown
+    attributes:
+      value: |
+        ## Environment
+
+  - type: input
+    id: cli-version
+    attributes:
+      label: CLI version
+      description: "Output of bl --version (use only X.Y.Z, without the bl prefix)"
+      placeholder: "1.2.1"
+    validations:
+      required: true
+
+  - type: input
+    id: skill-version
+    attributes:
+      label: Skill version (optional)
+      description: "metadata.version from the installed bailian-cli skill, if applicable"
+      placeholder: "1.2.1"
+
+  - type: input
+    id: node-version
+    attributes:
+      label: Node version
+      description: "Output of node --version"
+      placeholder: "v22.12.0"
+    validations:
+      required: true
+
+  - type: input
+    id: os
+    attributes:
+      label: OS
+      description: "e.g. darwin 24.5.0, Ubuntu 22.04"
+      placeholder: "darwin 24.5.0"
+    validations:
+      required: true
+
+  - type: dropdown
+    id: region
+    attributes:
+      label: Region
+      description: "From bl auth status or bl config show"
+      options:
+        - cn
+        - us
+        - intl
+        - unknown
+    validations:
+      required: true
+
+  - type: markdown
+    attributes:
+      value: |
+        ## Reproduction
+
+  - type: textarea
+    id: reproduce-command
+    attributes:
+      label: Command to reproduce
+      description: "Exact command that failed. Redact --api-key, sk-..., and sensitive prompts."
+      render: shell
+      placeholder: |
+        bl video generate --prompt "sunset" --download out.mp4 --verbose
+    validations:
+      required: true
+
+  - type: textarea
+    id: expected
+    attributes:
+      label: Expected behavior
+      description: What should have happened?
+    validations:
+      required: true
+
+  - type: textarea
+    id: actual
+    attributes:
+      label: Actual behavior
+      description: What happened instead?
+    validations:
+      required: true
+
+  - type: markdown
+    attributes:
+      value: |
+        ## Error output
+
+        Paste stderr as printed by `bl`. Include `Request ID` when present — it helps us trace logs.
+
+  - type: textarea
+    id: full-output
+    attributes:
+      label: Full output
+      description: Error, Hint, Status, Request ID, Exit code, etc.
+      render: shell
+      placeholder: |
+        Error:       Generation completed but no images returned.
+        Hint:        ...
+        Status:      HTTP 200 (...)
+        Request ID:  ...
+        Exit code:   1
+    validations:
+      required: true
+
+  - type: textarea
+    id: json-error
+    attributes:
+      label: JSON error (optional)
+      description: "Re-run with --output json and paste the error object if available"
+      render: json
+      placeholder: |
+        {
+          "error": {
+            "code": 1,
+            "message": "...",
+            "http_status": 200,
+            "api_code": "...",
+            "request_id": "..."
+          }
+        }
+
+  - type: markdown
+    attributes:
+      value: |
+        ## Troubleshooting already tried
+
+  - type: checkboxes
+    id: already-tried
+    attributes:
+      label: Already tried
+      options:
+        - label: "bl update and skill version aligned with CLI"
+        - label: "bl auth status OK for this command"
+        - label: "Different network / region — still reproduces"
+
+  - type: markdown
+    attributes:
+      value: |
+        ## Additional context
+
+  - type: dropdown
+    id: frequency
+    attributes:
+      label: How often does this happen?
+      options:
+        - Always
+        - Intermittent
+        - Once
+    validations:
+      required: true
+
+  - type: dropdown
+    id: invoked-via
+    attributes:
+      label: How was bl invoked?
+      options:
+        - Terminal (manual)
+        - Agent (Cursor, Claude, etc.)
+        - CI / script
+        - Other
+    validations:
+      required: true
+
+  - type: textarea
+    id: notes
+    attributes:
+      label: Notes (optional)
+      description: Anything else that might help — related issues, screenshots, minimal repro repo, etc.
@@ -33,6 +33,7 @@ tools/generated
 .claude/worktrees/
 .claude/settings.json
 .claude/settings.local.json
+.claude/scheduled_tasks.lock
 .cursor/
 .qwen/
 .playwright-mcp/
@@ -0,0 +1 @@
+24.16.0
@@ -93,6 +93,18 @@ CLI 只为「自己能权威解释的错误」发出语义化信号,服务端的

 不要扮演服务端错误的翻译官——我们没有最新的错误码体系认知,二次包装只会撒谎(详见 `docs/agents/error-hint-change.md` 中的反面 case)。

+### 4. Console Gateway 命令必须声明 console 全局 flags
+
+如果新命令使用了 `callConsoleGateway`，必须在 `options` 中添加以下三个全局 flag 的说明，以便 `--help` 中展示：
+
+```ts
+{ flag: "--console-region <region>", description: "Console region" },
+{ flag: "--console-site <site>", description: "Console site: domestic, international" },
+{ flag: "--console-switch-agent <uid>", description: "Switch agent UID", type: "number" },
+```
+
+这些 flag 已在 `GLOBAL_OPTIONS`（`packages/core/src/types/command.ts`）中注册，由 `loadConfig` 写入 `config.consoleRegion` / `config.consoleSite` / `config.consoleSwitchAgent`，`callConsoleGateway` 自动读取——命令无需手动提取或传递。
+
 ## 完成改动后的快速验证

 ```sh
@@ -6,6 +6,110 @@ The format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/), and

 [中文版](CHANGELOG.zh.md) · [README](README.md) · [Contributing](CONTRIBUTING.md)

+## [1.4.2] - 2026-06-24
+
+### Added
+
+- `bl omni --list-voices` prints the built-in output voices (ID, name, description, language) and exits without needing an API key. The built-in voice table is expanded from 6 to 17 voices, including dialect voices such as Dylan, Sunny, and Kiki.
+
+### Changed
+
+- `bl omni` default `--voice` is now `Tina` (previously `Cherry`). The `--voice` help points at `--list-voices` instead of listing every option inline.
+- `bl speech synthesize --list-voices` and its missing-`--voice` hint now include a link to the official CosyVoice voice documentation.
+- Agent skill setup guidance now covers console site selection (`--console-site domestic` / `international`) for console login and gateway commands.
+
+### Fixed
+
+- `bl speech synthesize` corrects the `cosyvoice-v3-flash` built-in voice ID from `longanhuan` to `longanhuan_v3`.
+
+## [1.4.1] - 2026-06-22
+
+### Changed
+
+- Video generation now defaults to the upgraded HappyHorse 1.1 model for better quality. The 1.0 models are still available via `--model`.
+- `bl update` now keeps the agent skill in sync across all your agent apps (Claude Code, Cursor, etc.), and refreshes it even when the CLI is already up to date.
+
+## [1.4.0] - 2026-06-17
+
+### Added
+
+- Console gateway now supports multiple regions and sites: `cn-beijing` and `ap-southeast-1`, each with domestic and international variants, plus `switchAgent` for delegated access.
+- New global flags `--console-region`, `--console-site`, and `--console-switch-agent`; `bl console call` also gains `--site` and `--switch-agent`.
+- `bl auth login --base-url <url>` to specify the base URL when logging in with an API key.
+- `bl omni` gains a `--voice` option (Chelsie, Cherry, Ethan, Serena, Sunny, Tina; default Cherry).
+
+### Changed
+
+- All user-facing CLI text is now standardized to English.
+- `bl advisor recommend` internal intent/ranking model upgraded from `qwen-turbo` to `qwen-flash`.
+- Cleaner JSON output for `usage`, `quota`, and `workspace` commands.
+- `base_url` from the config file now takes priority over the `DASHSCOPE_BASE_URL` environment variable.
+- `bl config show` now displays all fields from `config.json`, with sensitive values masked.
+
+### Removed
+
+- The legacy `region` config field and its related options.
+- Invalid leftover code for the removed `model list` command.
+
+### Fixed
+
+- When the console session is not logged in or has expired, the CLI now shows a clear sign-in prompt instead of a generic gateway error.
+- Corrected `--resolution` / `--ratio` / `--duration` flag descriptions for `bl video` commands.
+
+## [1.3.3] - 2026-06-16
+
+### Changed
+
+- `bl knowledge retrieve --help` now clearly indicates that `--api-key` is the recommended authentication method; AK/SK flags are explicitly marked as deprecated with guidance to use `--api-key` instead.
+
+### Added
+
+- `notes` field for command definitions — commands can now include contextual notes (auth requirements, deprecation notices, etc.) that are displayed in both `--help` output and the generated reference docs.
+
+## [1.3.2] - 2026-06-12
+
+### Fixed
+
+- Fixed `bl omni --audio` always returning HTTP 400 (#54); audio inputs are now understood correctly.
+
+## [1.3.1] - 2026-06-12
+
+### Fixed
+
+- `bl` now honors `HTTP_PROXY` / `HTTPS_PROXY` / `NO_PROXY` environment variables (#35). Node's built-in `fetch` (undici) ignores proxy env vars by default, causing `ECONNRESET` for users behind a VPN or corporate proxy. A global proxy dispatcher is now installed at startup when these variables are set, and the `ECONNRESET` error hint points to `export HTTPS_PROXY=http://127.0.0.1:<port>`.
+
+## [1.3.0] - 2026-06-10
+
+### Added
+
+- `bl knowledge retrieve` now supports API-Key authentication (DashScope gateway), in addition to AK/SK. API-Key is auto-detected and preferred when available.
+- New retrieval options: `--dense-similarity-top-k`, `--sparse-similarity-top-k`, `--rerank-model`, `--rerank-mode`, `--rerank-instruct` — supported on both API-Key and AK/SK paths.
+- `DashScopeKnowledgeRetrieveRequest` / `DashScopeKnowledgeRetrieveResponse` types and `knowledgeRetrieveEndpoint` added to `bailian-cli-core`.
+- Comprehensive E2E tests for knowledge retrieve covering both auth paths, dry-run, rerank flags, and error cases.
+
+- `bl usage` command group:
+  - `bl usage free` — query free-tier quota for all models (or a specific model with `--model`).
+  - `bl usage freetier` — enable (`--on`) or disable (`--off`) auto-stop for free-tier models.
+  - `bl usage stats` — query model usage statistics (requires `--workspace-id`).
+- `bl quota` command group:
+  - `bl quota list` — view model RPM/TPM rate limits (filter with `--model`, show all with `--all`).
+  - `bl quota check` — check current RPM/TPM usage against rate limits.
+  - `bl quota history` — view quota change history with pagination.
+  - `bl quota request` — request a temporary quota increase for a model.
+- `bl workspace list` — list all workspaces with region and endpoint details.
+
+### Changed
+
+- Credential resolution priority: explicit API-Key → explicit AK/SK flags → auto-detected API-Key → fallback AK/SK from config/env.
+- `--workspace-id` is now only required for AK/SK auth, no longer mandatory for API-Key mode.
+- `--top-k` deprecated in favor of `--rerank-top-n`; emits a warning and maps to `--rerank-top-n` when used.
+- `--access-key-id` / `--access-key-secret` flags marked as deprecated (API-Key is recommended).
+- API Key and console links updated to direct key management pages across all docs.
+
+### Fixed
+
+- `--rerank` flag in AK/SK path now correctly sets `EnableReranking` instead of the non-functional `Rerank: true` boolean.
+
 ## [1.2.1] - 2026-06-09

 ### Changed
@@ -6,6 +6,109 @@

 [English](CHANGELOG.md) · [README](README.zh.md) · [参与贡献](CONTRIBUTING.zh.md)

+## [1.4.2] - 2026-06-24
+
+### 新增
+
+- `bl omni --list-voices` 无需 API key 即可打印内置输出音色列表(ID、名称、描述、语言)并退出。内置音色表从 6 个扩展到 17 个,新增 Dylan、Sunny、Kiki 等方言音色。
+
+### 变更
+
+- `bl omni` 默认 `--voice` 改为 `Tina`(原为 `Cherry`)。`--voice` 帮助文案改为指向 `--list-voices`,不再内联列出全部音色。
+- `bl speech synthesize --list-voices` 输出及缺少 `--voice` 时的提示中,新增官方 CosyVoice 音色文档链接。
+- Agent skill 配置指引新增 console 站点选择说明(`--console-site domestic` / `international`),适用于 console 登录与网关类命令。
+
+### 修复
+
+- `bl speech synthesize` 修正 `cosyvoice-v3-flash` 内置音色 ID,由 `longanhuan` 改为 `longanhuan_v3`。
+
+## [1.4.1] - 2026-06-22
+
+### 变更
+
+- 视频生成默认升级到 HappyHorse 1.1 模型,画面质量更佳。如需使用 1.0 模型,可通过 `--model` 指定。
+- `bl update` 现在会把 agent skill 同步更新到所有 agent 应用(Claude Code、Cursor 等),即使 CLI 已是最新版本也会刷新 skill。
+
+## [1.4.0] - 2026-06-17
+
+### 新增
+
+- 控制台网关支持多 region 与多站点：`cn-beijing` 与 `ap-southeast-1`，各含国内站 / 国际站变体，并新增 `switchAgent` 委托访问。
+- 新增全局标志 `--console-region`、`--console-site`、`--console-switch-agent`；`bl console call` 另外新增 `--site` 与 `--switch-agent`。
+- `bl auth login --base-url <url>`：使用 API Key 登录时可指定 base URL。
+- `bl omni` 新增 `--voice` 选项（Chelsie、Cherry、Ethan、Serena、Sunny、Tina，默认 Cherry）。
+
+### 变更
+
+- 所有面向用户的 CLI 文案统一为英文。
+- `bl advisor recommend` 内部意图 / 排序模型由 `qwen-turbo` 升级为 `qwen-flash`。
+- 优化 `usage`、`quota`、`workspace` 命令的 JSON 输出。
+- 配置文件中的 `base_url` 现在优先级高于环境变量 `DASHSCOPE_BASE_URL`。
+- `bl config show` 现在展示 `config.json` 中的全部字段（敏感值已脱敏）。
+
+### 移除
+
+- 移除遗留的 `region` 配置字段及其相关选项。
+- 清理 `model list` 命令移除后遗留的无效代码。
+
+### 修复
+
+- 当控制台会话未登录或已过期时，CLI 现在会给出明确的登录提示，不再是笼统的网关错误。
+- 修正 `bl video` 命令 `--resolution` / `--ratio` / `--duration` 的帮助文案。
+
+## [1.3.3] - 2026-06-16
+
+### 变更
+
+- `bl knowledge retrieve --help` 现在明确指出 `--api-key` 是推荐的鉴权方式；AK/SK 相关选项已标注废弃并引导用户使用 `--api-key`。
+
+### 新增
+
+- 命令定义新增 `notes` 字段 — 命令可以附带上下文说明（鉴权要求、废弃提示等），同时展示在 `--help` 输出和生成的命令手册中。
+
+## [1.3.2] - 2026-06-12
+
+### 修复
+
+- 修复 `bl omni --audio` 始终返回 HTTP 400 的问题（#54），音频输入现已能正常理解。
+
+## [1.3.1] - 2026-06-12
+
+### 修复
+
+- `bl` 现在会读取 `HTTP_PROXY` / `HTTPS_PROXY` / `NO_PROXY` 环境变量(#35)。Node 内置的 `fetch`(undici)默认忽略代理环境变量,导致 VPN 或公司代理下出现 `ECONNRESET`。现已在启动时根据这些变量安装全局代理 dispatcher,并在 `ECONNRESET` 报错提示中给出 `export HTTPS_PROXY=http://127.0.0.1:<port>` 的指引。
+
+## [1.3.0] - 2026-06-11
+
+### 新增
+
+- `bl usage` 命令组：
+  - `bl usage free` — 查询所有模型的免费额度（可通过 `--model` 指定模型）。
+  - `bl usage freetier` — 启用（`--on`）或禁用（`--off`）免费额度模型的自动停服。
+  - `bl usage stats` — 查询模型用量统计（需指定 `--workspace-id`）。
+- `bl quota` 命令组：
+  - `bl quota list` — 查看模型 RPM/TPM 速率限制（支持 `--model` 过滤，`--all` 展示全部）。
+  - `bl quota check` — 查看当前 RPM/TPM 用量与速率限制。
+  - `bl quota history` — 查看配额变更记录，支持分页。
+  - `bl quota request` — 申请模型临时配额提升。
+- `bl workspace list` — 列出所有业务空间，包含地域和 endpoint 信息。
+- `bl knowledge retrieve` 新增 API-Key 鉴权（DashScope 网关），与原有 AK/SK 并存，可用时自动优先使用 API-Key。
+- 新增检索参数：`--dense-similarity-top-k`、`--sparse-similarity-top-k`、`--rerank-model`、`--rerank-mode`、`--rerank-instruct`，API-Key 与 AK/SK 两条链路均支持。
+- `bailian-cli-core` 新增 `DashScopeKnowledgeRetrieveRequest` / `DashScopeKnowledgeRetrieveResponse` 类型及 `knowledgeRetrieveEndpoint` 端点。
+- 知识库检索全面 E2E 测试，覆盖两种鉴权路径、dry-run、rerank 参数及错误场景。
+
+### 变更
+
+- 凭据解析优先级：显式 API-Key → 显式 AK/SK flag → 自动检测 API-Key → 回退至配置/环境变量中的 AK/SK。
+- `--workspace-id` 仅在 AK/SK 鉴权时必填，API-Key 模式下不再强制要求。
+- `--top-k` 标记为废弃，改用 `--rerank-top-n`；使用时输出警告并自动映射。
+- `--access-key-id` / `--access-key-secret` 标记为废弃（推荐使用 API-Key）。
+- 全部文档中的 API Key 和控制台链接更新为直达密钥管理页面。
+
+### 修复
+
+- AK/SK 链路 `--rerank` 现在正确设置 `EnableReranking`，而非之前无效的 `Rerank: true` 布尔值。
+
 ## [1.2.1] - 2026-06-09

 ### 变更
@@ -102,7 +102,7 @@ bl auth status --output json
 bl text chat --message "ping" --non-interactive --output json
 ```

-若失败：根据 stderr / JSON 中的 `hint` 或 `message` 排查（网络、Key 无效、region 等）。全局 region：`--region cn|us|intl`，默认 `cn`。
+若失败：根据 stderr / JSON 中的 `hint` 或 `message` 排查（网络、Key 无效、`base_url` 等）。DashScope 端点：使用 `--base-url` / `bl config set --key base_url` / `DASHSCOPE_BASE_URL`，默认中国大陆 `https://dashscope.aliyuncs.com`。

 ---

@@ -9,7 +9,7 @@
 [![TypeScript](https://img.shields.io/badge/TypeScript-strict-3178c6)](https://www.typescriptlang.org)
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE)

-[Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [中文文档](https://github.com/modelstudioai/cli/blob/main/README.zh.md) · [API Documentation](https://help.aliyun.com/zh/model-studio/) · [Get API Key](https://bailian.console.aliyun.com/cli?source_channel=key_github&)
+[Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [中文文档](https://github.com/modelstudioai/cli/blob/main/README.zh.md) · [API Documentation](https://help.aliyun.com/zh/model-studio/) · [Get API Key](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key)

 ---

@@ -27,15 +27,18 @@ Equip your AI Agent out-of-the-box with these capabilities, composable across co
 - **Text chat** — Qwen3.7-max: major gains in agentic coding, frontend coding, and vibe coding
 - **Multimodal (Omni)** — Full omni-modal support across text + image + audio + video
 - **Image generation & editing** — Qwen-Image 2.0: pro text rendering, photorealism, strong semantic adherence, multi-image composition
- **Video generation & editing** — HappyHorse-1.0 series: text-/image-/reference-to-video and natural-language video editing (up to 9-image reference)
+- **Video generation & editing** — happyhorse-1.1 series: text-/image-/reference-to-video and natural-language video editing (up to 9-image reference)
 - **Speech synthesis & recognition** — CosyVoice streaming TTS, voice cloning from 5–20s samples; FunAudio-ASR covers 30 languages including 7 Chinese dialects and 20+ Mandarin accents
 - **Image & video understanding** — Qwen-VL: long-form video analysis, chart/document parsing, visual reasoning, multilingual OCR
+
+> **Note:** The features below are currently available only to China site (aliyun.com) account holders and are not yet supported for international / global site accounts.
+
 - **Knowledge base & memory** — Multimodal RAG retrieval and cross-session memory for personalized, coherent dialogue
 - **App calls** — Invoke agents and workflows already published on Aliyun Model Studio
 - **MCP integration** — Orchestrate Bailian MCP servers: list services, inspect tools, and invoke any tool directly from the terminal
 - **Web search** — Real-time internet retrieval for up-to-date, accurate answers
 - **Model recommendation** — Describe your scenario and get best-fit model suggestions; supports scoped search, model comparison, and alternative discovery
- **Console capabilities** — Browse Bailian apps (`app list`) and check free-tier quota (`usage free`)
+- **Console capabilities** — Browse Bailian apps (`app list`), check free-tier quota (`usage free`), view model usage statistics (`usage stats`), manage workspaces (`workspace list`), and manage rate limits (`quota list/request/check/history`)
 - **Local file auto-upload** — Every URL parameter accepts a local path; uploaded to free temp storage with 48-hour validity

 ## Showcase: One-Sentence Cinematic Video
@@ -51,7 +54,7 @@ Equip your AI Agent out-of-the-box with these capabilities, composable across co
 A complete **2-minute, 16:9 cinematic short film** — produced end-to-end from a single natural-language sentence, with **zero manual editing**. This showcase demonstrates how an AI Agent can compose a multi-step creative pipeline by orchestrating three primitives:

 - **[Qwen Code](https://github.com/QwenLM/qwen-code)** — the agentic coding model that interprets the user's intent and drives the workflow
- **[Aliyun Model Studio CLI](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)** — invokes **HappyHorse 1.0**, Aliyun Model Studio's text-/image-/reference-to-video generation model
+- **[Aliyun Model Studio CLI](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)** — invokes **HappyHorse 1.1**, Aliyun Model Studio's text-/image-/reference-to-video generation model
 - **[spark-video Skill](https://github.com/JohnKeating1997/spark-video)** — handles scene decomposition, storyboarding, shot continuity, and final stitching

 ### The single prompt
@@ -64,7 +67,7 @@ A complete **2-minute, 16:9 cinematic short film** — produced end-to-end from

 1. **Qwen Code** parses the request, plans the narrative beats, and decides which tools to call.
 2. The **spark-video Skill** breaks the story into shots, writes per-shot prompts, and enforces visual continuity (characters, lighting, palette, lens language).
-3. **`bl video generate`** dispatches each shot to **HappyHorse 1.0** in parallel.
+3. **`bl video generate`** dispatches each shot to **HappyHorse 1.1** in parallel.
 4. The skill stitches all clips back together into a single 16:9 / ~2-min deliverable.

 No timeline scrubbing. No frame-by-frame editing. Just one sentence → one video.
@@ -108,9 +111,22 @@ bl advisor recommend --message "qwen-max vs deepseek-v3 for code generation"
 # Browser login (required for console capability commands)
 bl auth login --console

-# Browse apps / free-tier quota
+# Browse apps / free-tier quota / usage statistics / workspaces
 bl app list
 bl usage free --model qwen3-max
+bl usage free --expiring 30                           # Quotas expiring within 30 days
+bl usage free --sort remaining                        # Sort by remaining % ascending
+bl usage stats --workspace-id <id>                    # Usage overview for a workspace
+bl usage stats --model qwen-turbo --workspace-id <id> # Per-model usage
+bl workspace list                                     # List all workspaces
+
+# Rate limit management
+bl quota list                                         # View RPM/TPM limits for all models
+bl quota list --model qwen3.6-plus                    # View limits for a specific model
+bl quota check                                        # Current usage vs rate limits
+bl quota check --model qwen3.6-plus --period 5        # Check usage over last 5 minutes
+bl quota request --model qwen3.6-plus --tpm 6000000   # Request a temporary TPM increase
+bl quota history                                      # View quota change history
 ```

 > More examples and scenarios: [Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)
@@ -119,7 +135,7 @@ bl usage free --model qwen3-max

 ### DashScope API Key

-Required for most commands. Get your key from the [DashScope Console](https://bailian.console.aliyun.com/cli?source_channel=key_github&).
+Required for most commands. Get your key from the [DashScope Console](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key).

 ```bash
 # Option 1: Environment variable
@@ -134,7 +150,7 @@ bl text chat --api-key sk-xxxxx --message "Hello"

 ### Console Login (OAuth)

-Required for console capability commands (`app list`, `usage free`). Opens the Bailian console in your browser to sign in.
+Required for console capability commands (`app list`, `usage free`, `usage stats`, `workspace list`, `quota list/request/check/history`). Opens the Bailian console in your browser to sign in.

 ```bash
 bl auth login --console
@@ -159,7 +175,7 @@ export BAILIAN_WORKSPACE_ID=ws-...
 bl config show

 # Set defaults
-bl config set --key region --value us
+bl config set --key base_url --value https://dashscope-us.aliyuncs.com
 bl config set --key default_text_model --value qwen-turbo
 bl config set --key timeout --value 600

@@ -171,14 +187,14 @@ Config file location: `~/.bailian/config.json`

 ## Links

-| Resource                     | URL                                                               |
-| :--------------------------- | :---------------------------------------------------------------- |
-| Aliyun Model Studio CLI Site | https://bailian.console.aliyun.com/cli?source_channel=cli_github& |
-| DashScope API Docs           | https://help.aliyun.com/zh/model-studio/                          |
-| Qwen Model List              | https://help.aliyun.com/zh/model-studio/getting-started/models    |
-| Aliyun Model Studio Console  | https://bailian.console.aliyun.com/                               |
-| Get API Key                  | https://bailian.console.aliyun.com/cli?source_channel=key_github& |
-| Get AccessKey                | https://ram.console.aliyun.com/manage/ak                          |
+| Resource                     | URL                                                                                       |
+| :--------------------------- | :---------------------------------------------------------------------------------------- |
+| Aliyun Model Studio CLI Site | https://bailian.console.aliyun.com/cli?source_channel=cli_github&                         |
+| DashScope API Docs           | https://help.aliyun.com/zh/model-studio/                                                  |
+| Qwen Model List              | https://help.aliyun.com/zh/model-studio/getting-started/models                            |
+| Aliyun Model Studio Console  | https://bailian.console.aliyun.com/?source_channel=cli_github                             |
+| Get API Key                  | https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key |
+| Get AccessKey                | https://ram.console.aliyun.com/manage/ak                                                  |

 ## Changelog

@@ -9,7 +9,7 @@
 [![TypeScript](https://img.shields.io/badge/TypeScript-strict-3178c6)](https://www.typescriptlang.org)
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE)

-[阿里云百炼 CLI 官方主页](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [English](https://github.com/modelstudioai/cli/blob/main/README.md) · [API 文档](https://help.aliyun.com/zh/model-studio/) · [获取 API Key](https://bailian.console.aliyun.com/cli?source_channel=key_github&)
+[阿里云百炼 CLI 官方主页](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [English](https://github.com/modelstudioai/cli/blob/main/README.md) · [API 文档](https://help.aliyun.com/zh/model-studio/) · [获取 API Key](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key)

 ---

@@ -27,15 +27,18 @@ _专为 AI Agent 打造，每个命令均可作为结构化工具调用。_
 - **文本对话** — Qwen3.7-max：Agentic coding、前端编程、Vibe coding 等能力显著增强
 - **全模态对话** — 文本 + 图像 + 音频 + 视频全模态支持
 - **图像生成与编辑** — Qwen-Image 2.0：专业文字渲染、真实质感、强语义遵循、多图合成
- **视频生成与编辑** — HappyHorse-1.0 系列，支持文生 / 图生 / 参考生（最多 9 张图参考）/ 自然语言视频编辑
+- **视频生成与编辑** — happyhorse-1.1 系列，支持文生 / 图生 / 参考生（最多 9 张图参考）/ 自然语言视频编辑
 - **语音合成与识别** — CosyVoice 实时流式合成，5-20s 样本即可克隆；FunAudio-ASR 覆盖 30 种语种，含汉语七大方言与 20+ 口音官话
 - **图像与视频理解** — Qwen-VL：长视频解析、复杂图表与文档识别、视觉推理、多语种 OCR
+
+> **注意：** 以下功能目前仅对中国站（aliyun.com）账号开放，国际站 / 全球站账号暂不支持。
+
 - **知识库与记忆库** — 多模态 RAG 检索 + 跨会话记忆，提供个性化连贯对话体验
 - **应用调用** — 调用已发布在阿里云百炼平台上的智能体与工作流应用
 - **MCP 集成** — 统一调度百炼 MCP 服务：列出服务、查看工具、直接在终端调用任意工具
 - **联网搜索** — 实时互联网信息检索，提升回答准确性及时效性
 - **模型推荐** — 描述你的场景，智能推荐最适合的模型；支持限定范围搜索、模型对比和替代发现
- **控制台能力** — 浏览百炼应用（`app list`），查询模型免费额度（`usage free`）
+- **控制台能力** — 浏览百炼应用（`app list`），查询模型免费额度（`usage free`），查看模型用量统计（`usage stats`），管理业务空间（`workspace list`），管理限流与提额（`quota list/request/check/history`）
 - **本地文件自动上传** — 所有 URL 参数同时支持本地路径，免费临时存储 48 小时

 ## 示例:一句话生成一部电影短片
@@ -51,7 +54,7 @@ _专为 AI Agent 打造，每个命令均可作为结构化工具调用。_
 一部完整的 **2 分钟、16:9 电影感短片** —— 由一句自然语言端到端生成,**全程零手动剪辑**。这个示例展示了 AI Agent 如何把三个基础能力编排成一条多步创作流水线:

 - **[Qwen Code](https://github.com/QwenLM/qwen-code)** —— Agentic coding 模型,解析用户意图、驱动整个工作流
- **[阿里云百炼 CLI](https://github.com/modelstudioai/cli/)** —— 调用 **HappyHorse 1.0**,百炼的文生/图生/参考生视频模型
+- **[阿里云百炼 CLI](https://github.com/modelstudioai/cli/)** —— 调用 **HappyHorse 1.1**,百炼的文生/图生/参考生视频模型
 - **[spark-video Skill](https://github.com/JohnKeating1997/spark-video)** —— 负责场景拆分、分镜设计、镜头连贯性和最终拼接

 ### 唯一的提示词
@@ -62,7 +65,7 @@ _专为 AI Agent 打造，每个命令均可作为结构化工具调用。_

 1. **Qwen Code** 解析需求、规划叙事节奏,决定要调用哪些工具。
 2. **spark-video Skill** 把故事拆成镜头、为每个镜头写提示词,并保证视觉连贯性(角色、光线、色调、镜头语言)。
-3. **`bl video generate`** 把每个镜头并行下发给 **HappyHorse 1.0**。
+3. **`bl video generate`** 把每个镜头并行下发给 **HappyHorse 1.1**。
 4. Skill 把所有片段拼成最终的 16:9 / 约 2 分钟成片。

 没有时间线拖拽,没有逐帧剪辑。一句话 → 一部短片。
@@ -103,9 +106,22 @@ bl advisor recommend --message "qwen-max 和 deepseek-v3 哪个更适合做代
 # 浏览器登录（控制台能力相关命令需要）
 bl auth login --console

-# 浏览应用 / 免费额度
+# 浏览应用 / 免费额度 / 用量统计 / 业务空间
 bl app list
 bl usage free --model qwen3-max
+bl usage free --expiring 30                           # 30 天内过期的额度
+bl usage free --sort remaining                        # 按剩余百分比升序排列
+bl usage stats --workspace-id <id>                    # 指定空间的用量概览
+bl usage stats --model qwen-turbo --workspace-id <id> # 指定模型用量
+bl workspace list                                     # 列出所有业务空间
+
+# 限流管理与提额
+bl quota list                                         # 查看所有模型的 RPM/TPM 限额
+bl quota list --model qwen3.6-plus                    # 查看指定模型限额
+bl quota check                                        # 查看当前用量 vs 限流阈值
+bl quota check --model qwen3.6-plus --period 5        # 查看最近 5 分钟用量
+bl quota request --model qwen3.6-plus --tpm 6000000   # 申请临时 TPM 提额
+bl quota history                                      # 查看提额历史记录
 ```

 > 更多案例与使用场景：[阿里云百炼 CLI 官方主页](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)
@@ -114,7 +130,7 @@ bl usage free --model qwen3-max

 ### DashScope API Key

-大部分命令均需要 API Key。前往 [DashScope 控制台](https://bailian.console.aliyun.com/cli?source_channel=key_github&) 获取。
+大部分命令均需要 API Key。前往 [DashScope 控制台](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key) 获取。

 ```bash
 # 方式一：环境变量
@@ -129,7 +145,7 @@ bl text chat --api-key sk-xxxxx --message "你好"

 ### 控制台登录（OAuth）

-控制台能力命令（`app list`、`usage free`）需要使用此登录方式。打开浏览器跳转百炼控制台完成登录。
+控制台能力命令（`app list`、`usage free`、`usage stats`、`workspace list`、`quota list/request/check/history`）需要使用此登录方式。打开浏览器跳转百炼控制台完成登录。

 ```bash
 bl auth login --console
@@ -154,7 +170,7 @@ export BAILIAN_WORKSPACE_ID=ws-...
 bl config show

 # 设置默认值
-bl config set --key region --value us
+bl config set --key base_url --value https://dashscope-us.aliyuncs.com
 bl config set --key default_text_model --value qwen-turbo
 bl config set --key timeout --value 600

@@ -166,14 +182,14 @@ bl update

 ## 相关链接

-| 资源                    | 地址                                                              |
-| :---------------------- | :---------------------------------------------------------------- |
-| 阿里云百炼 CLI 官方主页 | https://bailian.console.aliyun.com/cli?source_channel=cli_github& |
-| DashScope API 文档      | https://help.aliyun.com/zh/model-studio/                          |
-| 通义千问模型列表        | https://help.aliyun.com/zh/model-studio/getting-started/models    |
-| 阿里云百炼控制台        | https://bailian.console.aliyun.com/                               |
-| 获取 API Key            | https://bailian.console.aliyun.com/cli?source_channel=key_github& |
-| 获取 AccessKey          | https://ram.console.aliyun.com/manage/ak                          |
+| 资源                    | 地址                                                                                      |
+| :---------------------- | :---------------------------------------------------------------------------------------- |
+| 阿里云百炼 CLI 官方主页 | https://bailian.console.aliyun.com/cli?source_channel=cli_github&                         |
+| DashScope API 文档      | https://help.aliyun.com/zh/model-studio/                                                  |
+| 通义千问模型列表        | https://help.aliyun.com/zh/model-studio/getting-started/models                            |
+| 阿里云百炼控制台        | https://bailian.console.aliyun.com/?source_channel=cli_github                             |
+| 获取 API Key            | https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key |
+| 获取 AccessKey          | https://ram.console.aliyun.com/manage/ak                                                  |

 ## 更新日志

@@ -78,9 +78,7 @@ flag 优先 ─→ config 文件 ─→ env var

 ### D. main 启动逻辑

- [ ] `packages/cli/src/main.ts:NO_AUTH_SETUP` 列表:
-  - 如果新增的命令"自己管鉴权或不需要鉴权",加进去绕开 ensureApiKey 拦截
-  - 当前清单以 `main.ts:NO_AUTH_SETUP` 为准
+- [ ] 若新增命令**自行处理鉴权**或**不应在入口触发默认 API key 引导**,在对应 `defineCommand` 上设 `skipDefaultApiKeySetup: true`(见 `packages/core/src/types/command.ts`;`packages/cli/src/main.ts` 在 `registry.resolve` 后读取 `command.skipDefaultApiKeySetup`)

 ### E. 错误文案

@@ -50,14 +50,14 @@ git diff --name-only <base>...<head>
 - [ ] **`package.json` 没破坏发布元数据**:`bin` / `exports` / `files` / `inlinedDependencies` 字段任何删除或改名都要单独评估
 - [ ] **公共依赖没被悄悄升级**:catalog / 根 lockfile 改动要列出来
 - [ ] **`package.json` version 没倒退**:目标分支已经更高时(如 main 1.0.3 vs head 1.0.0-beta.1),手动对齐版本号,不要被 head 覆盖
- [ ] **全局表没冲突**:`registry.ts`、`NO_AUTH_SETUP`(`packages/cli/src/main.ts`)、`ExitCode` 三个全局表新增项不和现有项冲突
+- [ ] **全局表没冲突**:`registry.ts`、`defineCommand` 的 `skipDefaultApiKeySetup`(见 `packages/core/src/types/command.ts`)、`ExitCode` 三处新增项不和现有项冲突

 ## 清单 B:用户透出(用户可见的新东西必看)

 - [ ] **新命令 / 新 flag** 已同步到用户面文档:
  - [README.md](README.md) + [README.zh.md](README.zh.md)(中英文都要,常漏 `_CN`)
  - (SKILL.md 已迁出本仓库,由 `npx add skills` 机制独立维护,不在本仓库 review 范围)
- [ ] **`bl <cmd> --help`** 文案完整:`description` / `examples` / `apiDocs` 都填了
+- [ ] **`bl <cmd> --help`** 文案完整:`description` / `examples` 都填了
 - [ ] **demo / quickstart**:用户可调用的新命令至少有一个示例
 - [ ] **行为变化的老命令**:在 commit message / CHANGELOG 注明用户感知的差异
 - [ ] **错误信息 / 提示文案**:面向用户的字符串通顺、双语(项目主体是中文场景)
@@ -80,7 +80,7 @@ git diff --name-only <base>...<head>
 解冲突要点(merge 时不要漏):
  - <冲突文件> + <字段/段落> + <怎么取舍>
  ↑ 放"合并那一刻才会出现"的细节,例如 package.json 的 files/scripts/devDependencies 各取并集、
-    NO_AUTH_SETUP 这种全局表两边都加项时不要丢一侧、pnpm-lock.yaml 直接 rm 后 pnpm install 重生等。
+    `skipDefaultApiKeySetup` 这类命令元数据两边都加项时不要丢一侧、pnpm-lock.yaml 直接 rm 后 pnpm install 重生等。
 建议修(可后置):
  - ...
 仅信息(无需动作,告知即可):
@@ -94,11 +94,11 @@ git diff --name-only <base>...<head>

 ## 常见漏点(基于历史踩坑)

-| 漏点                                                                           | 后果                                                                          |
-| ------------------------------------------------------------------------------ | ----------------------------------------------------------------------------- |
-| `pnpm-workspace.yaml` 把 `packages/*` 收窄成显式列表                           | 合并后目标分支的新子包不再被 workspace 识别,`pnpm install` 看似正常但子包失联 |
-| 源分支 version 比目标分支低,直接 merge 覆盖                                    | npm 上版本号回退,latest tag 错乱                                              |
-| `registry.ts` 注册新命令但忘了 [README](README.md) / [README.zh](README.zh.md) | 用户完全感知不到新功能                                                        |
-| 共享 util 重构(抽公共函数)只改了一处调用方                                     | 其它调用方静默走旧分支,行为分裂                                               |
-| `NO_AUTH_SETUP` 加了不该免登录的命令                                           | 安全风险,用户没登录也能调付费 API                                             |
-| `NO_AUTH_SETUP` / `registry.ts` 这类全局表两边都加项,解冲突时被合掉一侧        | 某个命令突然要求登录 / 某个新命令注册丢失,编译能过、回归不易察觉              |
+| 漏点                                                                            | 后果                                                                          |
+| ------------------------------------------------------------------------------- | ----------------------------------------------------------------------------- |
+| `pnpm-workspace.yaml` 把 `packages/*` 收窄成显式列表                            | 合并后目标分支的新子包不再被 workspace 识别,`pnpm install` 看似正常但子包失联 |
+| 源分支 version 比目标分支低,直接 merge 覆盖                                     | npm 上版本号回退,latest tag 错乱                                              |
+| `registry.ts` 注册新命令但忘了 [README](README.md) / [README.zh](README.zh.md)  | 用户完全感知不到新功能                                                        |
+| 共享 util 重构(抽公共函数)只改了一处调用方                                      | 其它调用方静默走旧分支,行为分裂                                               |
+| 不该跳过默认 API key 引导的命令误设 `skipDefaultApiKeySetup: true`              | 安全风险,用户没配置 key 也能调付费 API                                        |
+| `catalog.ts` / `skipDefaultApiKeySetup` 这类元数据两边都加项,解冲突时被合掉一侧 | 某个命令突然要求登录 / 某个新命令注册丢失,编译能过、回归不易察觉              |
@@ -27,7 +27,7 @@
 命令元数据以 **`catalog.ts` 为单一登记处**;`registry.ts` 只负责解析与打印 help,不再内嵌命令表或手写 Resources 列表。

 ```
-commands/<...>.ts   defineCommand({ name, description, usage, options, examples, apiDocs?, run })
+commands/<...>.ts   defineCommand({ name, description, usage, options, examples, run })
        ↓
 commands/catalog.ts   export const commands: Record<string, Command>
        ↓
@@ -53,8 +53,7 @@ registry.ts  main.ts      tools/generate-reference.ts   export-schema.ts
  - 增删 `import xxx from "./.../xxx.ts"`
  - 在 `export const commands` 里增删 `"<group> <action>": xxx`(key 与 `defineCommand({ name })` 一致)
 - [ ] **不要**在 `registry.ts` 里重复登记命令(已从 catalog 读取)
- [ ] 命令需在 `bl help` / `reference/` 展示 API 文档链接时,在 `defineCommand` 里设 `apiDocs`(相对路径);help 与 reference 均从此字段生成
- [ ] 如果命令需要鉴权之外的特殊路径,看 `packages/cli/src/main.ts` 的 `NO_AUTH_SETUP`
+- [ ] 如果命令需要跳过入口的默认 DashScope API key 引导(`ensureApiKey`),在对应 `defineCommand` 上设 `skipDefaultApiKeySetup: true`(字段定义见 `packages/core/src/types/command.ts`;`main.ts` 根据已解析的 `command` 读取)
 - [ ] **`config/export-schema.ts`**: 若新命令不适合作为 agent tool,评估是否加入 `SKIP_PREFIXES`;该文件在 `run()` 内 `import("../catalog.ts")`,勿顶层 import catalog 以免循环依赖

 ### B. 文档层
@@ -67,6 +67,12 @@ node tools/release/publish-channel.mjs --channel test --dry-run
 - [ ] `packages/cli/package.json` 和 `packages/core/package.json` 已升到目标版本
 - [ ] pre-release 格式正确（`1.0.0-beta.0` / `1.0.0-rc.1`，**不要直接用 `1.0.0` 当 beta**）

+### CHANGELOG（仅 stable）
+
+- [ ] `CHANGELOG.md` 和 `CHANGELOG.zh.md` 都已新增目标版本条目，中英文一一对应
+- [ ] 分类标题用 Keep a Changelog 规范的 `Added` / `Changed` / `Deprecated` / `Removed` / `Fixed` / `Security`（中文版对应 `新增` / `变更` / `已弃用` / `已移除` / `修复` / `安全`），**不要自创 `Improved` / `优化` 等规范外分类**
+- [ ] 条目日期与发版日期一致
+
 ### 用户面文档

 - [ ] `README.md` / `README.zh.md` 的 Quick Start 命令仍能跑通
@@ -81,11 +87,12 @@ node tools/release/publish-channel.mjs --channel test --dry-run

 ## 常见漏点（基于历史踩坑）

-| 漏点                                                  | 后果                                               |
-| ----------------------------------------------------- | -------------------------------------------------- |
-| cli 升版号但 core 没升                                | check.mjs 会拦下                                   |
-| `1.0.0` 当 beta 直接发                                | 占了 `latest` tag，所有用户被强升，撤回成本极高    |
-| README 写的 bin 名实际 `package.json.bin` 没注册      | 用户复制命令报 `command not found`                 |
-| Node 徽章 `>=18`、engines `>=22.12` 不一致            | 用户在 Node 18 上 `npm i` 被 engine 警告或直接失败 |
-| npm Trusted Publisher 的 workflow filename 改了没同步 | OIDC 匹配不上，publish 报 404                      |
-| CI 用 Node 22（npm 10）跑 publish                     | npm 10 不支持 OIDC token 交换，publish 报 404      |
+| 漏点                                                     | 后果                                               |
+| -------------------------------------------------------- | -------------------------------------------------- |
+| cli 升版号但 core 没升                                   | check.mjs 会拦下                                   |
+| 发版漏更 CHANGELOG，或分类写成规范外的 `优化`/`Improved` | 用户看不到本次变更，分类与历史不一致               |
+| `1.0.0` 当 beta 直接发                                   | 占了 `latest` tag，所有用户被强升，撤回成本极高    |
+| README 写的 bin 名实际 `package.json.bin` 没注册         | 用户复制命令报 `command not found`                 |
+| Node 徽章 `>=18`、engines `>=22.12` 不一致               | 用户在 Node 18 上 `npm i` 被 engine 警告或直接失败 |
+| npm Trusted Publisher 的 workflow filename 改了没同步    | OIDC 匹配不上，publish 报 404                      |
+| CI 用 Node 22（npm 10）跑 publish                        | npm 10 不支持 OIDC token 交换，publish 报 404      |
@@ -0,0 +1,438 @@
+# 模型训练 + 数据集 + 部署：最小闭环 CLI 设计
+
+> 目标：一个 Qwen 文本模型 SFT 训练、数据集上传、模型部署的端到端最小链路。
+
+---
+
+## 一、命令概览
+
+| 优先级 | 命令                                | 映射 API                                      | 用途                            |
+| ------ | ----------------------------------- | --------------------------------------------- | ------------------------------- |
+| P0     | `bl dataset upload <path>`          | `POST /api/v1/files`                          | 上传训练数据(含本地格式校验)    |
+| P0     | `bl finetune create`                | `POST /api/v1/fine-tunes`                     | 创建 SFT 训练任务(预填默认超参) |
+| P0     | `bl finetune status <job_id>`       | `GET /api/v1/fine-tunes/{job_id}`             | 查询训练状态                    |
+| P0     | `bl deploy create`                  | `POST /api/v1/deployments`                    | 部署训练好的模型                |
+| P1     | `bl finetune logs <job_id>`         | `GET /api/v1/fine-tunes/{job_id}/logs`        | 拉取训练日志                    |
+| P1     | `bl finetune checkpoints <job_id>`  | `GET /api/v1/fine-tunes/{job_id}/checkpoints` | 查看/挑选 Checkpoint            |
+| P1     | `bl deploy status <deployed_model>` | `GET /api/v1/deployments/{deployed_model}`    | 查询部署状态                    |
+| P1     | `bl deploy delete <deployed_model>` | `DELETE /api/v1/deployments/{deployed_model}` | 下线部署                        |
+| P1     | `bl infer --model <deployed_model>` | 复用 `text chat` 通路                         | 调用已部署模型                  |
+
+---
+
+## 二、P0 命令详细设计
+
+### 2.1 `bl dataset upload`
+
+**定位：** 上传训练数据文件到百炼平台，获取 `file_id` 供训练任务引用。
+
+#### CLI 签名
+
+```
+bl dataset upload <path> [--purpose fine-tune] [--validate] [--no-validate]
+```
+
+| Flag            | 必填 | 默认值      | 说明                           |
+| --------------- | ---- | ----------- | ------------------------------ |
+| `<path>`        | 是   | —           | 本地文件路径（.jsonl 或 .zip） |
+| `--purpose`     | 否   | `fine-tune` | 文件用途标签                   |
+| `--validate`    | 否   | `true`      | 上传前执行本地格式校验         |
+| `--no-validate` | 否   | —           | 跳过本地校验                   |
+
+#### 本地格式校验规则（提交前拦截）
+
+校验逻辑在 `packages/core` 实现（纯函数），CLI 调用后展示错误：
+
+1. **文件格式检查**：仅允许 `.jsonl` 和 `.zip`（zip 内根目录必须有 `data.jsonl`）
+2. **JSONL 逐行校验**：
+   - 每行可被 `JSON.parse`
+   - 顶层必须包含 `messages` 数组
+   - `messages` 中每项必须包含 `role`（枚举：`system` | `user` | `assistant`）和 `content`（非空字符串）
+   - 至少包含一条 `user` + 一条 `assistant` 消息
+3. **数量校验**：SFT 训练至少需要上千条数据（给出 warning 而非 hard fail，阈值建议 ≥ 10 条 hard fail）
+4. **文件体积**：≤ 300MB
+
+#### 校验失败输出示例
+
+```
+✗ Validation failed:
+
+  Line 3: missing "messages" field
+  Line 7: role "bot" is not valid (expected: system | user | assistant)
+  Line 12: "content" is empty string
+
+Fix 3 errors above and retry.
+```
+
+#### API 调用
+
+```
+POST https://dashscope.aliyuncs.com/api/v1/files
+Content-Type: multipart/form-data
+Authorization: Bearer <api-key>
+
+Body:
+  files: <binary>
+  purpose: "fine-tune"
+
+Response 200:
+{
+  "id": "file-xxxx",
+  "bytes": 12345,
+  "filename": "train.jsonl",
+  "purpose": "fine-tune",
+  "created_at": 1700000000
+}
+```
+
+#### 输出
+
+- 默认 text：`✓ Uploaded file-xxxx (12.3 KB) — use this ID in bl finetune create`
+- `--output json`：完整 response body
+- `--quiet`：仅输出 `file-xxxx`
+
+---
+
+### 2.2 `bl finetune create`
+
+**定位：** 创建一个 SFT 训练任务。核心设计原则——**预填合理默认超参 + 提交前二次确认**，降低 OOM/超参不合理导致的训练失败率。
+
+#### CLI 签名
+
+```
+bl finetune create --model <model> --data <file_id> [hyperparams...]
+```
+
+| Flag                | 必填 | 默认值       | 说明                                         |
+| ------------------- | ---- | ------------ | -------------------------------------------- |
+| `--model`           | 是   | —            | 基座模型（如 `qwen3-8b`, `qwen3-14b`）       |
+| `--data`            | 是   | —            | 训练数据 file_id（bl dataset upload 返回值） |
+| `--validation-data` | 否   | —            | 验证数据 file_id                             |
+| `--epochs`          | 否   | 3            | 训练轮次 (n_epochs)                          |
+| `--batch-size`      | 否   | 按模型自动选 | 批大小                                       |
+| `--lr`              | 否   | 按模型自动选 | 学习率 (learning_rate_multiplier)            |
+| `--warmup-ratio`    | 否   | 0.1          | warmup 比例                                  |
+| `--suffix`          | 否   | —            | 输出模型后缀名                               |
+| `--yes` / `-y`      | 否   | —            | 跳过确认直接提交                             |
+
+#### 预填默认超参策略
+
+| 基座模型   | batch_size | lr_multiplier | n_epochs | 备注             |
+| ---------- | ---------- | ------------- | -------- | ---------------- |
+| qwen3-8b   | 4          | 1e-5          | 3        | 小模型可大 batch |
+| qwen3-14b  | 2          | 5e-6          | 3        | 中模型防 OOM     |
+| qwen3-32b+ | 1          | 2e-6          | 2        | 大模型保守设置   |
+
+> 以上为建议默认值，用户显式传参时覆盖。具体映射表在 `packages/core/src/finetune/defaults.ts` 维护。
+
+#### 提交前交互确认
+
+非 `--yes` 模式下，显示任务摘要等待确认：
+
+```
+┌─ Fine-tune Job Summary ──────────────────────┐
+│ Model:        qwen3-8b                        │
+│ Training:     file-abc123 (2,048 samples)     │
+│ Validation:   (none)                          │
+│ Epochs:       3                               │
+│ Batch size:   4                               │
+│ LR:           1e-5                            │
+│ Warmup:       0.1                             │
+│ Suffix:       my-assistant                    │
+│                                               │
+│ Estimated cost: ~¥XX (based on token count)   │
+└───────────────────────────────────────────────┘
+Proceed? [Y/n]
+```
+
+#### API 调用
+
+```
+POST https://dashscope.aliyuncs.com/api/v1/fine-tunes
+Authorization: Bearer <api-key>
+Content-Type: application/json
+
+{
+  "model": "qwen3-8b",
+  "training_file_ids": ["file-abc123"],
+  "validation_file_ids": [],
+  "hyper_parameters": {
+    "n_epochs": 3,
+    "batch_size": 4,
+    "learning_rate": "1e-5",
+    "warmup_ratio": 0.1
+  },
+  "suffix": "my-assistant"
+}
+
+Response 200:
+{
+  "job_id": "ft-xxxx",
+  "status": "PENDING",
+  "model": "qwen3-8b",
+  "created_at": "2025-01-01T00:00:00Z",
+  "training_file_ids": ["file-abc123"],
+  "hyper_parameters": {...},
+  "trained_model": null
+}
+```
+
+#### 输出
+
+- text：`✓ Fine-tune job ft-xxxx created (PENDING). Track with: bl finetune status ft-xxxx`
+- json：完整 response body
+- quiet：`ft-xxxx`
+
+---
+
+### 2.3 `bl finetune status`
+
+**定位：** 查询训练任务状态，支持 `--wait` 轮询模式。
+
+#### CLI 签名
+
+```
+bl finetune status <job_id> [--wait] [--interval <seconds>]
+```
+
+| Flag         | 必填 | 默认值 | 说明             |
+| ------------ | ---- | ------ | ---------------- |
+| `<job_id>`   | 是   | —      | 任务 ID          |
+| `--wait`     | 否   | —      | 持续轮询直到终态 |
+| `--interval` | 否   | 30     | 轮询间隔(秒)     |
+
+#### 状态机
+
+```
+PENDING → RUNNING → SUCCEEDED
+                  ↘ FAILED
+```
+
+#### 输出（text 模式）
+
+单次查询：
+
+```
+Job:     ft-xxxx
+Status:  RUNNING (elapsed 12m)
+Model:   qwen3-8b
+Output:  (pending)
+```
+
+`--wait` 模式（spinner + 实时刷新）：
+
+```
+⠋ ft-xxxx RUNNING [14:32 elapsed]
+✓ ft-xxxx SUCCEEDED — trained model: qwen3-8b:ft-xxxx-20250101
+  Deploy with: bl deploy create --model qwen3-8b:ft-xxxx-20250101
+```
+
+失败时：
+
+```
+✗ ft-xxxx FAILED
+  Error: OutOfMemory — try reducing --batch-size or using a smaller model
+```
+
+---
+
+### 2.4 `bl deploy create`
+
+**定位：** 将训练好的模型（或 checkpoint）部署为可调用的推理服务。
+
+#### CLI 签名
+
+```
+bl deploy create --model <model_name> [--plan <plan>] [--capacity <n>]
+```
+
+| Flag         | 必填 | 默认值     | 说明                                            |
+| ------------ | ---- | ---------- | ----------------------------------------------- |
+| `--model`    | 是   | —          | 待部署模型名称（finetune 产出的 trained_model） |
+| `--plan`     | 否   | `standard` | 部署方案                                        |
+| `--capacity` | 否   | 依 plan    | 并发容量                                        |
+| `--wait`     | 否   | —          | 等待部署就绪                                    |
+
+#### API 调用
+
+```
+POST https://dashscope.aliyuncs.com/api/v1/deployments
+Authorization: Bearer <api-key>
+Content-Type: application/json
+
+{
+  "model_name": "qwen3-8b:ft-xxxx-20250101",
+  "plan": "standard",
+  "capacity": 2
+}
+
+Response 200:
+{
+  "deployed_model": "qwen3-8b-ft-xxxx",
+  "model_name": "qwen3-8b:ft-xxxx-20250101",
+  "status": "PENDING",
+  "created_at": "..."
+}
+```
+
+#### 输出
+
+```
+✓ Deployment created: qwen3-8b-ft-xxxx (PENDING)
+  Once RUNNING, call with: bl text chat --model qwen3-8b-ft-xxxx
+  Check status: bl deploy status qwen3-8b-ft-xxxx
+```
+
+---
+
+## 三、P1 命令简要设计
+
+### 3.1 `bl finetune logs <job_id>`
+
+流式输出训练日志，支持 `--follow`（类似 `tail -f`）。输出 loss/step/epoch 信息。
+
+### 3.2 `bl finetune checkpoints <job_id>`
+
+列出可选 checkpoint（step, loss, eval metrics），支持 `--output json` 供脚本使用。可配合 `bl deploy create --model <checkpoint_model>` 部署指定 checkpoint。
+
+### 3.3 `bl deploy status <deployed_model>`
+
+查询部署状态及资源信息（PENDING → RUNNING → STOPPED/FAILED）。
+
+### 3.4 `bl deploy delete <deployed_model>`
+
+下线部署。需部署处于 RUNNING/STOPPED/FAILED 状态。交互确认或 `--yes` 跳过。
+
+### 3.5 `bl infer --model <deployed_model>`
+
+实际可复用已有 `bl text chat --model <deployed_model>` 通路，作为别名/快捷方式。P1 考虑是否有独立存在必要。
+
+---
+
+## 四、代码架构方案
+
+按照 monorepo 分层约定（core 纯逻辑 / cli 是 UI）：
+
+### packages/core 新增模块
+
+```
+packages/core/src/
+├── finetune/
+│   ├── index.ts          # re-export
+│   ├── api.ts            # createFineTune, getFineTune, getFineTuneLogs, getCheckpoints
+│   ├── defaults.ts       # 模型 → 默认超参映射表
+│   └── types.ts          # FineTuneJob, HyperParameters, CheckpointInfo 类型
+├── dataset/
+│   ├── index.ts
+│   ├── upload.ts         # uploadDataset (multipart)
+│   ├── validate.ts       # validateJsonl (纯函数,逐行校验)
+│   └── types.ts          # DatasetFile, ValidationError 类型
+└── deploy/
+    ├── index.ts
+    ├── api.ts            # createDeployment, getDeployment, deleteDeployment
+    └── types.ts          # Deployment, DeploymentStatus 类型
+```
+
+### packages/cli 新增命令
+
+```
+packages/cli/src/commands/
+├── dataset/
+│   └── upload.ts         # bl dataset upload
+├── finetune/
+│   ├── create.ts         # bl finetune create
+│   ├── status.ts         # bl finetune status
+│   ├── logs.ts           # bl finetune logs
+│   └── checkpoints.ts    # bl finetune checkpoints
+└── deploy/
+    ├── create.ts         # bl deploy create
+    ├── status.ts         # bl deploy status
+    └── delete.ts         # bl deploy delete
+```
+
+---
+
+## 五、关键设计决策
+
+### 5.1 数据格式校验放在 CLI 侧（提交前拦截）
+
+训练失败 TOP 原因中"数据格式错误"占比高。与其等服务端 10 分钟后返回 FAILED，不如 CLI 本地秒级校验：
+
+- **validate.ts** 是纯函数，接收 ReadableStream/Buffer，返回 `ValidationError[]`
+- CLI 在 `dataset upload` 默认执行校验，`--no-validate` 允许跳过
+- 未来可扩展为独立命令 `bl dataset validate <path>`
+
+### 5.2 超参预填 + 确认而非强制
+
+- core 维护 `defaults.ts` 映射：`model → { batch_size, lr, epochs }`
+- CLI `finetune create` 未指定超参时自动填入
+- 提交前展示完整参数面板（非 --yes 模式），避免"我以为用了默认但其实没传"
+
+### 5.3 费用感知（P1+）
+
+- 图像/语音/视频训练费用远高于文本。MVP 阶段（Qwen 文本 SFT）费用可控
+- 后续扩展多模态时，在 confirm panel 中强化费用估算提示
+- `bl quota check` 已存在，可在 `finetune create` 内部集成余额预检
+
+### 5.4 `bl infer` 是否独立存在
+
+建议 P1 阶段**不新增** `bl infer`，而是让 `bl text chat --model <deployed_model>` 直接工作。部署完成后的引导文案中指明这个用法即可。减少命令膨胀。
+
+---
+
+## 六、最小闭环用户操作流
+
+```bash
+# 1. 准备数据 → 上传（含校验）
+bl dataset upload ./train.jsonl
+# ✓ Uploaded file-abc123 (5.2 MB)
+
+# 2. 创建训练任务（自动预填超参）
+bl finetune create --model qwen3-8b --data file-abc123
+# Shows summary panel → confirm → ✓ Job ft-xxxx created
+
+# 3. 等待训练完成
+bl finetune status ft-xxxx --wait
+# ⠋ RUNNING [23:15] → ✓ SUCCEEDED: qwen3-8b:ft-xxxx-20250601
+
+# 4. 部署模型
+bl deploy create --model qwen3-8b:ft-xxxx-20250601 --wait
+# ✓ Deployed: qwen3-8b-ft-xxxx (RUNNING)
+
+# 5. 调用模型
+bl text chat --model qwen3-8b-ft-xxxx "你好，介绍一下你自己"
+# (正常推理输出)
+```
+
+---
+
+## 七、实现顺序建议
+
+```
+Phase 1 (P0 — 最小闭环):
+  core: dataset/validate.ts → dataset/upload.ts → finetune/api.ts → deploy/api.ts
+  cli:  dataset upload → finetune create → finetune status → deploy create
+  测试: 单元测试 validate.ts + e2e dry-run + 真实 API 端到端一次
+
+Phase 2 (P1 — 可观测性):
+  finetune logs → finetune checkpoints → deploy status → deploy delete
+  费用估算集成
+
+Phase 3 (后续):
+  bl dataset validate (独立命令)
+  bl dataset list (查看已上传)
+  bl finetune list (查看历史任务)
+  多模态 SFT 支持（图像/视频数据格式校验扩展）
+```
+
+---
+
+## 八、风险与 TODO
+
+| 风险点            | 影响              | 缓解措施                                      |
+| ----------------- | ----------------- | --------------------------------------------- |
+| OOM 训练失败      | 用户浪费时间/金钱 | 保守默认超参 + batch_size 自适应模型大小      |
+| 数据格式错误      | 训练启动后才失败  | 本地校验拦截，启动秒级反馈                    |
+| 部署等待时间长    | 用户困惑          | `--wait` + 预估时间提示                       |
+| 费用超预期        | 账号欠费          | confirm panel 预估费用（P1 集成 quota check） |
+| API endpoint 变动 | 调用失败          | 端点集中管理在 core/client/endpoints.ts       |
@@ -9,7 +9,7 @@
 [![TypeScript](https://img.shields.io/badge/TypeScript-strict-3178c6)](https://www.typescriptlang.org)
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE)

-[Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [中文文档](https://github.com/modelstudioai/cli/blob/main/README.zh.md) · [API Documentation](https://help.aliyun.com/zh/model-studio/) · [Get API Key](https://bailian.console.aliyun.com/cli?source_channel=key_github&)
+[Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [中文文档](https://github.com/modelstudioai/cli/blob/main/README.zh.md) · [API Documentation](https://help.aliyun.com/zh/model-studio/) · [Get API Key](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key)

 ---

@@ -27,15 +27,18 @@ Equip your AI Agent out-of-the-box with these capabilities, composable across co
 - **Text chat** — Qwen3.7-max: major gains in agentic coding, frontend coding, and vibe coding
 - **Multimodal (Omni)** — Full omni-modal support across text + image + audio + video
 - **Image generation & editing** — Qwen-Image 2.0: pro text rendering, photorealism, strong semantic adherence, multi-image composition
- **Video generation & editing** — HappyHorse-1.0 series: text-/image-/reference-to-video and natural-language video editing (up to 9-image reference)
+- **Video generation & editing** — happyhorse-1.1 series: text-/image-/reference-to-video and natural-language video editing (up to 9-image reference)
 - **Speech synthesis & recognition** — CosyVoice streaming TTS, voice cloning from 5–20s samples; FunAudio-ASR covers 30 languages including 7 Chinese dialects and 20+ Mandarin accents
 - **Image & video understanding** — Qwen-VL: long-form video analysis, chart/document parsing, visual reasoning, multilingual OCR
+
+> **Note:** The features below are currently available only to China site (aliyun.com) account holders and are not yet supported for international / global site accounts.
+
 - **Knowledge base & memory** — Multimodal RAG retrieval and cross-session memory for personalized, coherent dialogue
 - **App calls** — Invoke agents and workflows already published on Aliyun Model Studio
 - **MCP integration** — Orchestrate Bailian MCP servers: list services, inspect tools, and invoke any tool directly from the terminal
 - **Web search** — Real-time internet retrieval for up-to-date, accurate answers
 - **Model recommendation** — Describe your scenario and get best-fit model suggestions; supports scoped search, model comparison, and alternative discovery
- **Console capabilities** — Browse Bailian apps (`app list`) and check free-tier quota (`usage free`)
+- **Console capabilities** — Browse Bailian apps (`app list`), check free-tier quota (`usage free`), view model usage statistics (`usage stats`), manage workspaces (`workspace list`), and manage rate limits (`quota list/request/check/history`)
 - **Local file auto-upload** — Every URL parameter accepts a local path; uploaded to free temp storage with 48-hour validity

 ## Showcase: One-Sentence Cinematic Video
@@ -51,7 +54,7 @@ Equip your AI Agent out-of-the-box with these capabilities, composable across co
 A complete **2-minute, 16:9 cinematic short film** — produced end-to-end from a single natural-language sentence, with **zero manual editing**. This showcase demonstrates how an AI Agent can compose a multi-step creative pipeline by orchestrating three primitives:

 - **[Qwen Code](https://github.com/QwenLM/qwen-code)** — the agentic coding model that interprets the user's intent and drives the workflow
- **[Aliyun Model Studio CLI](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)** — invokes **HappyHorse 1.0**, Aliyun Model Studio's text-/image-/reference-to-video generation model
+- **[Aliyun Model Studio CLI](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)** — invokes **HappyHorse 1.1**, Aliyun Model Studio's text-/image-/reference-to-video generation model
 - **[spark-video Skill](https://github.com/JohnKeating1997/spark-video)** — handles scene decomposition, storyboarding, shot continuity, and final stitching

 ### The single prompt
@@ -64,7 +67,7 @@ A complete **2-minute, 16:9 cinematic short film** — produced end-to-end from

 1. **Qwen Code** parses the request, plans the narrative beats, and decides which tools to call.
 2. The **spark-video Skill** breaks the story into shots, writes per-shot prompts, and enforces visual continuity (characters, lighting, palette, lens language).
-3. **`bl video generate`** dispatches each shot to **HappyHorse 1.0** in parallel.
+3. **`bl video generate`** dispatches each shot to **HappyHorse 1.1** in parallel.
 4. The skill stitches all clips back together into a single 16:9 / ~2-min deliverable.

 No timeline scrubbing. No frame-by-frame editing. Just one sentence → one video.
@@ -108,9 +111,22 @@ bl advisor recommend --message "qwen-max vs deepseek-v3 for code generation"
 # Browser login (required for console capability commands)
 bl auth login --console

-# Browse apps / free-tier quota
+# Browse apps / free-tier quota / usage statistics / workspaces
 bl app list
 bl usage free --model qwen3-max
+bl usage free --expiring 30                           # Quotas expiring within 30 days
+bl usage free --sort remaining                        # Sort by remaining % ascending
+bl usage stats --workspace-id <id>                    # Usage overview for a workspace
+bl usage stats --model qwen-turbo --workspace-id <id> # Per-model usage
+bl workspace list                                     # List all workspaces
+
+# Rate limit management
+bl quota list                                         # View RPM/TPM limits for all models
+bl quota list --model qwen3.6-plus                    # View limits for a specific model
+bl quota check                                        # Current usage vs rate limits
+bl quota check --model qwen3.6-plus --period 5        # Check usage over last 5 minutes
+bl quota request --model qwen3.6-plus --tpm 6000000   # Request a temporary TPM increase
+bl quota history                                      # View quota change history
 ```

 > More examples and scenarios: [Aliyun Model Studio CLI Site](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)
@@ -119,7 +135,7 @@ bl usage free --model qwen3-max

 ### DashScope API Key

-Required for most commands. Get your key from the [DashScope Console](https://bailian.console.aliyun.com/cli?source_channel=key_github&).
+Required for most commands. Get your key from the [DashScope Console](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key).

 ```bash
 # Option 1: Environment variable
@@ -134,7 +150,7 @@ bl text chat --api-key sk-xxxxx --message "Hello"

 ### Console Login (OAuth)

-Required for console capability commands (`app list`, `usage free`). Opens the Bailian console in your browser to sign in.
+Required for console capability commands (`app list`, `usage free`, `usage stats`, `workspace list`, `quota list/request/check/history`). Opens the Bailian console in your browser to sign in.

 ```bash
 bl auth login --console
@@ -159,7 +175,7 @@ export BAILIAN_WORKSPACE_ID=ws-...
 bl config show

 # Set defaults
-bl config set --key region --value us
+bl config set --key base_url --value https://dashscope-us.aliyuncs.com
 bl config set --key default_text_model --value qwen-turbo
 bl config set --key timeout --value 600

@@ -171,14 +187,14 @@ Config file location: `~/.bailian/config.json`

 ## Links

-| Resource                     | URL                                                               |
-| :--------------------------- | :---------------------------------------------------------------- |
-| Aliyun Model Studio CLI Site | https://bailian.console.aliyun.com/cli?source_channel=cli_github& |
-| DashScope API Docs           | https://help.aliyun.com/zh/model-studio/                          |
-| Qwen Model List              | https://help.aliyun.com/zh/model-studio/getting-started/models    |
-| Aliyun Model Studio Console  | https://bailian.console.aliyun.com/                               |
-| Get API Key                  | https://bailian.console.aliyun.com/cli?source_channel=key_github& |
-| Get AccessKey                | https://ram.console.aliyun.com/manage/ak                          |
+| Resource                     | URL                                                                                       |
+| :--------------------------- | :---------------------------------------------------------------------------------------- |
+| Aliyun Model Studio CLI Site | https://bailian.console.aliyun.com/cli?source_channel=cli_github&                         |
+| DashScope API Docs           | https://help.aliyun.com/zh/model-studio/                                                  |
+| Qwen Model List              | https://help.aliyun.com/zh/model-studio/getting-started/models                            |
+| Aliyun Model Studio Console  | https://bailian.console.aliyun.com/?source_channel=cli_github                             |
+| Get API Key                  | https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key |
+| Get AccessKey                | https://ram.console.aliyun.com/manage/ak                                                  |

 ## Changelog

@@ -9,7 +9,7 @@
 [![TypeScript](https://img.shields.io/badge/TypeScript-strict-3178c6)](https://www.typescriptlang.org)
 [![License](https://img.shields.io/badge/license-Apache%202.0-blue)](LICENSE)

-[阿里云百炼 CLI 官方主页](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [English](https://github.com/modelstudioai/cli/blob/main/README.md) · [API 文档](https://help.aliyun.com/zh/model-studio/) · [获取 API Key](https://bailian.console.aliyun.com/cli?source_channel=key_github&)
+[阿里云百炼 CLI 官方主页](https://bailian.console.aliyun.com/cli?source_channel=cli_github&) · [English](https://github.com/modelstudioai/cli/blob/main/README.md) · [API 文档](https://help.aliyun.com/zh/model-studio/) · [获取 API Key](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key)

 ---

@@ -27,15 +27,18 @@ _专为 AI Agent 打造，每个命令均可作为结构化工具调用。_
 - **文本对话** — Qwen3.7-max：Agentic coding、前端编程、Vibe coding 等能力显著增强
 - **全模态对话** — 文本 + 图像 + 音频 + 视频全模态支持
 - **图像生成与编辑** — Qwen-Image 2.0：专业文字渲染、真实质感、强语义遵循、多图合成
- **视频生成与编辑** — HappyHorse-1.0 系列，支持文生 / 图生 / 参考生（最多 9 张图参考）/ 自然语言视频编辑
+- **视频生成与编辑** — happyhorse-1.1 系列，支持文生 / 图生 / 参考生（最多 9 张图参考）/ 自然语言视频编辑
 - **语音合成与识别** — CosyVoice 实时流式合成，5-20s 样本即可克隆；FunAudio-ASR 覆盖 30 种语种，含汉语七大方言与 20+ 口音官话
 - **图像与视频理解** — Qwen-VL：长视频解析、复杂图表与文档识别、视觉推理、多语种 OCR
+
+> **注意：** 以下功能目前仅对中国站（aliyun.com）账号开放，国际站 / 全球站账号暂不支持。
+
 - **知识库与记忆库** — 多模态 RAG 检索 + 跨会话记忆，提供个性化连贯对话体验
 - **应用调用** — 调用已发布在阿里云百炼平台上的智能体与工作流应用
 - **MCP 集成** — 统一调度百炼 MCP 服务：列出服务、查看工具、直接在终端调用任意工具
 - **联网搜索** — 实时互联网信息检索，提升回答准确性及时效性
 - **模型推荐** — 描述你的场景，智能推荐最适合的模型；支持限定范围搜索、模型对比和替代发现
- **控制台能力** — 浏览百炼应用（`app list`），查询模型免费额度（`usage free`）
+- **控制台能力** — 浏览百炼应用（`app list`），查询模型免费额度（`usage free`），查看模型用量统计（`usage stats`），管理业务空间（`workspace list`），管理限流与提额（`quota list/request/check/history`）
 - **本地文件自动上传** — 所有 URL 参数同时支持本地路径，免费临时存储 48 小时

 ## 示例:一句话生成一部电影短片
@@ -51,7 +54,7 @@ _专为 AI Agent 打造，每个命令均可作为结构化工具调用。_
 一部完整的 **2 分钟、16:9 电影感短片** —— 由一句自然语言端到端生成,**全程零手动剪辑**。这个示例展示了 AI Agent 如何把三个基础能力编排成一条多步创作流水线:

 - **[Qwen Code](https://github.com/QwenLM/qwen-code)** —— Agentic coding 模型,解析用户意图、驱动整个工作流
- **[阿里云百炼 CLI](https://github.com/modelstudioai/cli/)** —— 调用 **HappyHorse 1.0**,百炼的文生/图生/参考生视频模型
+- **[阿里云百炼 CLI](https://github.com/modelstudioai/cli/)** —— 调用 **HappyHorse 1.1**,百炼的文生/图生/参考生视频模型
 - **[spark-video Skill](https://github.com/JohnKeating1997/spark-video)** —— 负责场景拆分、分镜设计、镜头连贯性和最终拼接

 ### 唯一的提示词
@@ -62,7 +65,7 @@ _专为 AI Agent 打造，每个命令均可作为结构化工具调用。_

 1. **Qwen Code** 解析需求、规划叙事节奏,决定要调用哪些工具。
 2. **spark-video Skill** 把故事拆成镜头、为每个镜头写提示词,并保证视觉连贯性(角色、光线、色调、镜头语言)。
-3. **`bl video generate`** 把每个镜头并行下发给 **HappyHorse 1.0**。
+3. **`bl video generate`** 把每个镜头并行下发给 **HappyHorse 1.1**。
 4. Skill 把所有片段拼成最终的 16:9 / 约 2 分钟成片。

 没有时间线拖拽,没有逐帧剪辑。一句话 → 一部短片。
@@ -103,9 +106,22 @@ bl advisor recommend --message "qwen-max 和 deepseek-v3 哪个更适合做代
 # 浏览器登录（控制台能力相关命令需要）
 bl auth login --console

-# 浏览应用 / 免费额度
+# 浏览应用 / 免费额度 / 用量统计 / 业务空间
 bl app list
 bl usage free --model qwen3-max
+bl usage free --expiring 30                           # 30 天内过期的额度
+bl usage free --sort remaining                        # 按剩余百分比升序排列
+bl usage stats --workspace-id <id>                    # 指定空间的用量概览
+bl usage stats --model qwen-turbo --workspace-id <id> # 指定模型用量
+bl workspace list                                     # 列出所有业务空间
+
+# 限流管理与提额
+bl quota list                                         # 查看所有模型的 RPM/TPM 限额
+bl quota list --model qwen3.6-plus                    # 查看指定模型限额
+bl quota check                                        # 查看当前用量 vs 限流阈值
+bl quota check --model qwen3.6-plus --period 5        # 查看最近 5 分钟用量
+bl quota request --model qwen3.6-plus --tpm 6000000   # 申请临时 TPM 提额
+bl quota history                                      # 查看提额历史记录
 ```

 > 更多案例与使用场景：[阿里云百炼 CLI 官方主页](https://bailian.console.aliyun.com/cli?source_channel=cli_github&)
@@ -114,7 +130,7 @@ bl usage free --model qwen3-max

 ### DashScope API Key

-大部分命令均需要 API Key。前往 [DashScope 控制台](https://bailian.console.aliyun.com/cli?source_channel=key_github&) 获取。
+大部分命令均需要 API Key。前往 [DashScope 控制台](https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key) 获取。

 ```bash
 # 方式一：环境变量
@@ -129,7 +145,7 @@ bl text chat --api-key sk-xxxxx --message "你好"

 ### 控制台登录（OAuth）

-控制台能力命令（`app list`、`usage free`）需要使用此登录方式。打开浏览器跳转百炼控制台完成登录。
+控制台能力命令（`app list`、`usage free`、`usage stats`、`workspace list`、`quota list/request/check/history`）需要使用此登录方式。打开浏览器跳转百炼控制台完成登录。

 ```bash
 bl auth login --console
@@ -154,7 +170,7 @@ export BAILIAN_WORKSPACE_ID=ws-...
 bl config show

 # 设置默认值
-bl config set --key region --value us
+bl config set --key base_url --value https://dashscope-us.aliyuncs.com
 bl config set --key default_text_model --value qwen-turbo
 bl config set --key timeout --value 600

@@ -166,14 +182,14 @@ bl update

 ## 相关链接

-| 资源                    | 地址                                                              |
-| :---------------------- | :---------------------------------------------------------------- |
-| 阿里云百炼 CLI 官方主页 | https://bailian.console.aliyun.com/cli?source_channel=cli_github& |
-| DashScope API 文档      | https://help.aliyun.com/zh/model-studio/                          |
-| 通义千问模型列表        | https://help.aliyun.com/zh/model-studio/getting-started/models    |
-| 阿里云百炼控制台        | https://bailian.console.aliyun.com/                               |
-| 获取 API Key            | https://bailian.console.aliyun.com/cli?source_channel=key_github& |
-| 获取 AccessKey          | https://ram.console.aliyun.com/manage/ak                          |
+| 资源                    | 地址                                                                                      |
+| :---------------------- | :---------------------------------------------------------------------------------------- |
+| 阿里云百炼 CLI 官方主页 | https://bailian.console.aliyun.com/cli?source_channel=cli_github&                         |
+| DashScope API 文档      | https://help.aliyun.com/zh/model-studio/                                                  |
+| 通义千问模型列表        | https://help.aliyun.com/zh/model-studio/getting-started/models                            |
+| 阿里云百炼控制台        | https://bailian.console.aliyun.com/?source_channel=cli_github                             |
+| 获取 API Key            | https://bailian.console.aliyun.com/cn-beijing/?source_channel=key_github&tab=app#/api-key |
+| 获取 AccessKey          | https://ram.console.aliyun.com/manage/ak                                                  |

 ## 更新日志

@@ -1,6 +1,6 @@
 {
  "name": "bailian-cli",
-  "version": "1.2.1",
+  "version": "1.4.2",
  "description": "CLI for Aliyun Model Studio (DashScope) AI Platform.",
  "keywords": [
    "agent",
@@ -46,7 +46,8 @@
  "dependencies": {
    "bailian-cli-core": "workspace:*",
    "boxen": "catalog:",
-    "chalk": "catalog:"
+    "chalk": "catalog:",
+    "undici": "catalog:"
  },
  "devDependencies": {
    "@clack/prompts": "^0.7.0",
@@ -29,41 +29,41 @@ function formatContextWindow(tokens: number): string {
 }

 const MODALITY_LABELS: Record<string, string> = {
-  Text: "文本",
-  Image: "图片",
-  Video: "视频",
-  Audio: "音频",
+  Text: "Text",
+  Image: "Image",
+  Video: "Video",
+  Audio: "Audio",
 };
 const CAPABILITY_LABELS: Record<string, string> = {
-  TG: "文本生成",
-  VU: "视觉理解",
-  IG: "图像生成",
-  VG: "视频生成",
-  TTS: "语音合成",
-  ASR: "语音识别",
-  Reasoning: "推理",
+  TG: "Text Gen",
+  VU: "Vision",
+  IG: "Image Gen",
+  VG: "Video Gen",
+  TTS: "Text-to-Speech",
+  ASR: "Speech-to-Text",
+  Reasoning: "Reasoning",
 };
 const BUDGET_LABELS: Record<string, string> = {
-  low: "低成本优先",
-  medium: "适中",
-  high: "高投入",
+  low: "Cost-Effective",
+  medium: "Balanced",
+  high: "High Investment",
 };
 const QUALITY_LABELS: Record<string, string> = {
-  flagship: "旗舰优先",
-  balanced: "均衡",
-  "cost-optimized": "性价比优先",
+  flagship: "Flagship",
+  balanced: "Balanced",
+  "cost-optimized": "Value",
 };
 const PREFERENCE_MODE_LABELS: Record<string, string> = {
-  scoped: "限定范围",
-  comparison: "对比评估",
-  alternative: "替代推荐",
+  scoped: "Scoped",
+  comparison: "Comparison",
+  alternative: "Alternative",
 };

 function formatIntentSummary(intent: IntentProfile, noColor: boolean): string {
  const colorize = noColor ? new Chalk({ level: 0 }) : chalk;

  const lines: string[] = [];
-  lines.push(colorize.cyan.bold("需求理解"));
+  lines.push(colorize.cyan.bold("Intent Analysis"));

  if (intent.taskSummary) {
    lines.push("");
@@ -72,7 +72,7 @@ function formatIntentSummary(intent: IntentProfile, noColor: boolean): string {

  if (intent.scenarioHints.length) {
    lines.push("");
-    lines.push(`${colorize.dim("场景特征")}  ${intent.scenarioHints.join(" · ")}`);
+    lines.push(`${colorize.dim("Scenario")}  ${intent.scenarioHints.join(" · ")}`);
  }

  const inputLabels = intent.inputModality.map((mod) => MODALITY_LABELS[mod] ?? mod);
@@ -80,40 +80,40 @@ function formatIntentSummary(intent: IntentProfile, noColor: boolean): string {
  if (inputLabels.length || outputLabels.length) {
    lines.push("");
    const parts: string[] = [];
-    if (inputLabels.length) parts.push(`${colorize.dim("输入")} ${inputLabels.join(", ")}`);
-    if (outputLabels.length) parts.push(`${colorize.dim("输出")} ${outputLabels.join(", ")}`);
+    if (inputLabels.length) parts.push(`${colorize.dim("Input")} ${inputLabels.join(", ")}`);
+    if (outputLabels.length) parts.push(`${colorize.dim("Output")} ${outputLabels.join(", ")}`);
    lines.push(parts.join("    "));
  }

  const capLabels = intent.requiredCapabilities.map((cap) => CAPABILITY_LABELS[cap] ?? cap);
  if (capLabels.length) {
-    lines.push(`${colorize.dim("所需能力")}  ${capLabels.join(", ")}`);
+    lines.push(`${colorize.dim("Capabilities")}  ${capLabels.join(", ")}`);
  }

  const budgetLabel = BUDGET_LABELS[intent.budget] ?? intent.budget;
  const qualityLabel = QUALITY_LABELS[intent.qualityPreference] ?? intent.qualityPreference;
  lines.push("");
  lines.push(
-    `${colorize.dim("预算倾向")} ${budgetLabel}    ${colorize.dim("质量偏好")} ${qualityLabel}`,
+    `${colorize.dim("Budget")} ${budgetLabel}    ${colorize.dim("Quality")} ${qualityLabel}`,
  );

  const preference = intent.modelPreference;
  if (preference && preference.mode !== "unconstrained") {
    lines.push("");
    const modeLabel = PREFERENCE_MODE_LABELS[preference.mode] ?? preference.mode;
-    const prefParts = [colorize.dim("推荐模式") + ` ${colorize.yellow(modeLabel)}`];
+    const prefParts = [colorize.dim("Mode") + ` ${colorize.yellow(modeLabel)}`];
    if (preference.targets?.length) {
-      prefParts.push(colorize.dim("目标") + ` ${preference.targets.join(", ")}`);
+      prefParts.push(colorize.dim("Targets") + ` ${preference.targets.join(", ")}`);
    }
    if (preference.excludes?.length) {
-      prefParts.push(colorize.dim("排除") + ` ${preference.excludes.join(", ")}`);
+      prefParts.push(colorize.dim("Excludes") + ` ${preference.excludes.join(", ")}`);
    }
    lines.push(prefParts.join("    "));
  }

  if (intent.segments?.length) {
    lines.push("");
-    lines.push(colorize.dim("任务拆解"));
+    lines.push(colorize.dim("Pipeline"));
    for (const [idx, segment] of intent.segments.entries()) {
      const outMods = segment.outputModality.map((mod) => MODALITY_LABELS[mod] ?? mod).join(", ");
      lines.push(
@@ -131,19 +131,19 @@ function formatIntentSummary(intent: IntentProfile, noColor: boolean): string {
  });
 }

-const RECOMMEND_LABELS = ["最佳推荐", "次优选择", "备选参考"];
+const RECOMMEND_LABELS = ["Best Pick", "Runner-Up", "Alternative"];

 function renderCard(rec: RecommendedModel, index: number, colorize: ChalkInstance): string {
  const labelColors = [colorize.green.bold, colorize.blue.bold, colorize.magenta.bold];
  const colorFn = labelColors[index] ?? colorize.white.bold;
-  const label = RECOMMEND_LABELS[index] ?? `推荐 #${index + 1}`;
+  const label = RECOMMEND_LABELS[index] ?? `#${index + 1}`;

  const lines: string[] = [];
-  lines.push(colorFn(`⬢ 推荐 #${index + 1} — ${label}`));
+  lines.push(colorFn(`⬢  #${index + 1} — ${label}`));
  lines.push("");
  lines.push(`${colorize.bold(rec.name)}  ${colorize.dim(`(${rec.model})`)}`);
  lines.push("");
-  lines.push(`${colorize.cyan("推荐理由")}  ${rec.reason}`);
+  lines.push(`${colorize.cyan("Why")}  ${rec.reason}`);

  if (rec.highlights.length) {
    lines.push("");
@@ -153,8 +153,8 @@ function renderCard(rec: RecommendedModel, index: number, colorize: ChalkInstanc
  }

  const meta: string[] = [];
-  if (rec.contextWindow) meta.push(`上下文 ${formatContextWindow(rec.contextWindow)}`);
-  if (rec.maxOutputTokens) meta.push(`最大输出 ${formatContextWindow(rec.maxOutputTokens)}`);
+  if (rec.contextWindow) meta.push(`Context ${formatContextWindow(rec.contextWindow)}`);
+  if (rec.maxOutputTokens) meta.push(`Max Output ${formatContextWindow(rec.maxOutputTokens)}`);
  if (meta.length) {
    lines.push("");
    lines.push(colorize.dim(meta.join("  ·  ")));
@@ -163,7 +163,7 @@ function renderCard(rec: RecommendedModel, index: number, colorize: ChalkInstanc
  const docLink = buildDocLink(rec.docUrl);
  if (docLink) {
    lines.push("");
-    lines.push(colorize.dim(`文档  ${docLink}`));
+    lines.push(colorize.dim(`Docs  ${docLink}`));
  }

  return boxen(lines.join("\n"), {
@@ -183,7 +183,7 @@ function formatSingleResult(results: RecommendedModel[], noColor: boolean): stri
 function formatPipelineResult(summary: string, steps: PipelineStep[], noColor: boolean): string {
  const colorize = noColor ? new Chalk({ level: 0 }) : chalk;
  const lines: string[] = [];
-  lines.push(`  ${colorize.yellow.bold("⚡ 组合方案")}  ${summary}`);
+  lines.push(`  ${colorize.yellow.bold("⚡ Pipeline")}  ${summary}`);

  for (const [stepIdx, { step, recommendations, warnings }] of steps.entries()) {
    lines.push("");
@@ -234,12 +234,12 @@ export default defineCommand({
    },
  ],
  examples: [
-    'bl advisor recommend --message "我要做一个能理解图片的客服机器人"',
-    'bl advisor recommend --message "做一个Agent自动根据用户意图生成动画片"',
-    'bl advisor recommend --message "法律合同审查，要求高精准度"',
-    'bl advisor recommend --message "做一个低成本高并发的在线客服" --output json',
-    'bl advisor recommend --message "长文本摘要" --dry-run',
-    "bl advisor recommend                                          # 交互式输入需求",
+    'bl advisor recommend --message "I need a visual-understanding chatbot"',
+    'bl advisor recommend --message "Build an Agent that auto-generates animations"',
+    'bl advisor recommend --message "Legal contract review, high precision required"',
+    'bl advisor recommend --message "Low-cost high-concurrency online customer service" --output json',
+    'bl advisor recommend --message "Long document summarization" --dry-run',
+    "bl advisor recommend                                          # Interactive input",
  ],
  async run(config: Config, flags: GlobalFlags) {
    const positional = ((flags as Record<string, unknown>)._positional as string[]) ?? [];
@@ -247,14 +247,14 @@ export default defineCommand({

    if (!userInput.trim()) {
      if (isInteractive({ nonInteractive: config.nonInteractive })) {
-        const hint = await promptText({ message: "描述你的需求:" });
+        const hint = await promptText({ message: "Describe your requirement:" });
        if (!hint) {
-          process.stderr.write("已取消。\n");
+          process.stderr.write("Cancelled.\n");
          process.exit(1);
        }
        userInput = hint;
      } else {
-        failIfMissing("message", 'bl advisor recommend "你的需求"');
+        failIfMissing("message", 'bl advisor recommend "your requirement"');
      }
    }

@@ -262,16 +262,16 @@ export default defineCommand({
    const format = detectOutputFormat(config.output);

    const modelsOptions: GetModelsOptions = {
-      onPrepareStart: () => process.stderr.write("初始化中...\n"),
+      onPrepareStart: () => process.stderr.write("Initializing model data...\n"),
    };
-    process.stderr.write("正在分析需求...\n");
+    process.stderr.write("Analyzing your request...\n");
    const [allModels, intent] = await Promise.all([
      getModels(config, modelsOptions),
      analyzeIntent(config, userInput),
    ]);

    if (intent.confidence === 0) {
-      process.stderr.write("需求分析超时，使用默认参数继续...\n");
+      process.stderr.write("Intent analysis timed out, using defaults...\n");
    } else {
      process.stderr.write("\n");
    }
@@ -297,7 +297,7 @@ export default defineCommand({
    }

    // Stage 3: LLM Ranking
-    const spinner = createSpinner("正在推荐最佳模型...");
+    const spinner = createSpinner("Recommending best models...");
    spinner.start();

    const result = await rankModels(config, candidates, intent, userInput, top);
@@ -305,12 +305,31 @@ export default defineCommand({
    spinner.stop();

    if (isEmptyResult(result)) {
-      emitBare("暂无满足该需求的模型。");
+      emitBare("No suitable models found for this request.");
      return;
    }

    if (format !== "text") {
-      emitResult(result, format);
+      emitResult(
+        {
+          intent: {
+            taskSummary: intent.taskSummary,
+            scenarioHints: intent.scenarioHints,
+            complexity: intent.complexity,
+            inputModality: intent.inputModality,
+            outputModality: intent.outputModality,
+            requiredCapabilities: intent.requiredCapabilities,
+            budget: intent.budget,
+            qualityPreference: intent.qualityPreference,
+            modelPreference:
+              intent.modelPreference?.mode !== "unconstrained" ? intent.modelPreference : undefined,
+            segments: intent.segments,
+          },
+          result,
+          candidates: candidates.length,
+        },
+        format,
+      );
      return;
    }

@@ -35,12 +35,12 @@ export default defineCommand({
    { flag: "--has-thoughts", description: "Show agent thinking process" },
  ],
  examples: [
-    'bl app call --app-id abc123 --prompt "你好"',
-    'bl app call --app-id abc123 --prompt "描述这张图片" --image https://example.com/photo.jpg',
-    'bl app call --app-id abc123 --prompt "分析图片" --image img1.jpg --image img2.jpg',
-    'bl app call --app-id abc123 --prompt "继续" --session-id sess_xxx --stream',
-    'bl app call --app-id abc123 --prompt "搜索资料" --pipeline-ids pipe1,pipe2',
-    'bl app call --app-id abc123 --prompt "开始" --biz-params \'{"key":"value"}\'',
+    'bl app call --app-id abc123 --prompt "Hello"',
+    'bl app call --app-id abc123 --prompt "Describe this image" --image https://example.com/photo.jpg',
+    'bl app call --app-id abc123 --prompt "Analyze the image" --image img1.jpg --image img2.jpg',
+    'bl app call --app-id abc123 --prompt "Continue" --session-id sess_xxx --stream',
+    'bl app call --app-id abc123 --prompt "Search for materials" --pipeline-ids pipe1,pipe2',
+    'bl app call --app-id abc123 --prompt "Start" --biz-params \'{"key":"value"}\'',
  ],
  async run(config: Config, flags: GlobalFlags) {
    const appId = flags.appId as string;
@@ -13,6 +13,7 @@ const APP_LIST_API = "zeldaEasy.broadscope-bailian.app-control.list";
 export default defineCommand({
  name: "app list",
  description: "List Bailian applications",
+  skipDefaultApiKeySetup: true,
  usage: "bl app list [flags]",
  options: [
    {
@@ -29,14 +30,20 @@ export default defineCommand({
      description: "Results per page (default: 30)",
      type: "number",
    },
+    { flag: "--console-region <region>", description: "Console region" },
    {
-      flag: "--region <region>",
-      description: "API region (default: cn-beijing)",
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
    },
  ],
  examples: [
    "bl app list",
-    "bl app list --name 客服",
+    "bl app list --name customer service",
    "bl app list --page 2 --page-size 10",
    "bl app list --output json",
  ],
@@ -44,7 +51,6 @@ export default defineCommand({
    const name = (flags.name as string) || "";
    const pageNo = (flags.page as number) || 1;
    const pageSize = (flags.pageSize as number) || 30;
-    const region = (flags.region as string) || "cn-beijing";
    const format = detectOutputFormat(config.output);

    const credential = await resolveConsoleGatewayCredential(config);
@@ -61,17 +67,13 @@ export default defineCommand({
    };

    if (config.dryRun) {
-      emitResult(
-        { api: APP_LIST_API, data, region, token: credential.token.slice(0, 8) + "..." },
-        format,
-      );
+      emitResult({ api: APP_LIST_API, data, token: credential.token.slice(0, 8) + "..." }, format);
      return;
    }

    const result = (await callConsoleGateway(config, credential.token, {
      api: APP_LIST_API,
      data,
-      region,
    })) as any;

    const list: unknown[] = result?.data?.DataV2?.data?.data?.list ?? [];
@@ -5,18 +5,24 @@ import http from "node:http";
 import {
  BailianError,
  ExitCode,
+  chatEndpoint,
  getConfigPath,
  readConfigFile,
+  requestJson,
  writeConfigFile,
+  type Config,
 } from "bailian-cli-core";

 const CONSOLE_LOGIN_TIMEOUT_MS = 15 * 60 * 1000;
 const MAX_AUTH_CALLBACK_BODY = 65536;

-const DEFAULT_CONSOLE_ORIGIN = "https://bailian.console.aliyun.com";
+const CONSOLE_ORIGINS: Record<string, string> = {
+  domestic: "https://bailian.console.aliyun.com",
+  international: "https://modelstudio.console.alibabacloud.com",
+};

-export function resolveConsoleOrigin(): string {
-  return process.env.BAILIAN_CONSOLE_ORIGIN || DEFAULT_CONSOLE_ORIGIN;
+export function resolveConsoleOrigin(site?: string): string {
+  return (site && CONSOLE_ORIGINS[site]) || CONSOLE_ORIGINS.domestic!;
 }

 function readBodyBounded(req: http.IncomingMessage): Promise<string> {
@@ -210,9 +216,76 @@ function parseApiKeyFromRawBody(raw: string, contentType: string): string | null
  return null;
 }

+type CallbackExtras = Pick<
+  CallbackCredentials,
+  "baseUrl" | "consoleSite" | "consoleRegion" | "consoleSwitchAgent" | "workspaceId"
+>;
+
+function stringField(o: Record<string, unknown>, ...keys: string[]): string | null {
+  for (const k of keys) {
+    const v = o[k];
+    if (typeof v === "string" && v.trim()) return v.trim();
+  }
+  return null;
+}
+
+function parseExtrasFromRawBody(raw: string, contentType: string): CallbackExtras {
+  const empty: CallbackExtras = {
+    baseUrl: null,
+    consoleSite: null,
+    consoleRegion: null,
+    consoleSwitchAgent: null,
+    workspaceId: null,
+  };
+  if (!raw.trim()) return empty;
+
+  let obj: Record<string, unknown> | null = null;
+
+  const ct = contentType.toLowerCase();
+  if (ct.includes("application/json") || ct.includes("text/json")) {
+    try {
+      const parsed = JSON.parse(raw.trim());
+      if (parsed && typeof parsed === "object" && !Array.isArray(parsed)) obj = parsed;
+    } catch {
+      /* */
+    }
+  }
+  if (!obj && ct.includes("application/x-www-form-urlencoded")) {
+    try {
+      const params = new URLSearchParams(raw.trim());
+      obj = Object.fromEntries(params);
+    } catch {
+      /* */
+    }
+  }
+  if (!obj) {
+    try {
+      const parsed = JSON.parse(raw.trim());
+      if (parsed && typeof parsed === "object" && !Array.isArray(parsed)) obj = parsed;
+    } catch {
+      /* */
+    }
+  }
+
+  if (!obj) return empty;
+
+  return {
+    baseUrl: stringField(obj, "base_url", "baseUrl"),
+    consoleSite: stringField(obj, "console_site", "consoleSite"),
+    consoleRegion: stringField(obj, "console_region", "consoleRegion"),
+    consoleSwitchAgent: stringField(obj, "console_switch_agent", "consoleSwitchAgent"),
+    workspaceId: stringField(obj, "workspace_id", "workspaceId"),
+  };
+}
+
 interface CallbackCredentials {
  accessToken: string | null;
  apiKey: string | null;
+  baseUrl: string | null;
+  consoleSite: string | null;
+  consoleRegion: string | null;
+  consoleSwitchAgent: string | null;
+  workspaceId: string | null;
 }

 async function extractCredentialsFromRequest(
@@ -222,12 +295,30 @@ async function extractCredentialsFromRequest(
  const accessTokenFromQuery =
    u.searchParams.get("access_token") ?? u.searchParams.get("accessToken");
  const apiKeyFromQuery = u.searchParams.get("api_key") ?? u.searchParams.get("apiKey");
+  const baseUrlFromQuery = u.searchParams.get("base_url") ?? u.searchParams.get("baseUrl");
+  const consoleSiteFromQuery =
+    u.searchParams.get("console_site") ?? u.searchParams.get("consoleSite");
+  const consoleRegionFromQuery =
+    u.searchParams.get("console_region") ?? u.searchParams.get("consoleRegion");
+  const consoleSwitchAgentFromQuery =
+    u.searchParams.get("console_switch_agent") ?? u.searchParams.get("consoleSwitchAgent");
+  const workspaceIdFromQuery =
+    u.searchParams.get("workspace_id") ?? u.searchParams.get("workspaceId");
+
+  const extras = {
+    baseUrl: baseUrlFromQuery?.trim() || null,
+    consoleSite: consoleSiteFromQuery?.trim() || null,
+    consoleRegion: consoleRegionFromQuery?.trim() || null,
+    consoleSwitchAgent: consoleSwitchAgentFromQuery?.trim() || null,
+    workspaceId: workspaceIdFromQuery?.trim() || null,
+  };

  const m = req.method ?? "GET";
  if (m !== "POST" && m !== "PUT" && m !== "PATCH") {
    return {
      accessToken: accessTokenFromQuery?.trim() || null,
      apiKey: apiKeyFromQuery?.trim() || null,
+      ...extras,
    };
  }

@@ -239,12 +330,24 @@ async function extractCredentialsFromRequest(
    return {
      accessToken: accessTokenFromQuery?.trim() || null,
      apiKey: apiKeyFromQuery?.trim() || null,
+      ...extras,
    };
  }

  const accessToken = accessTokenFromQuery?.trim() || parseAccessTokenFromRawBody(raw, contentType);
  const apiKey = apiKeyFromQuery?.trim() || parseApiKeyFromRawBody(raw, contentType);
-  return { accessToken, apiKey };
+
+  const bodyExtras = parseExtrasFromRawBody(raw, contentType);
+
+  return {
+    accessToken,
+    apiKey,
+    baseUrl: extras.baseUrl || bodyExtras.baseUrl,
+    consoleSite: extras.consoleSite || bodyExtras.consoleSite,
+    consoleRegion: extras.consoleRegion || bodyExtras.consoleRegion,
+    consoleSwitchAgent: extras.consoleSwitchAgent || bodyExtras.consoleSwitchAgent,
+    workspaceId: extras.workspaceId || bodyExtras.workspaceId,
+  };
 }

 function listenServerOnFreeLocalPort(server: http.Server): Promise<number> {
@@ -276,9 +379,69 @@ function openInBrowser(url: string): Promise<void> {
  });
 }

+const RETRY_DELAY_BASE_MS = 500;
+
+function canRetry(err: unknown): boolean {
+  if (err instanceof BailianError) {
+    if (err.exitCode === ExitCode.NETWORK || err.exitCode === ExitCode.TIMEOUT) return true;
+    const status = err.api?.httpStatus;
+    return status === 401 || (status !== undefined && status >= 500);
+  }
+  if (err instanceof Error) {
+    return (
+      err.name === "AbortError" ||
+      err.name === "TimeoutError" ||
+      err.message.includes("timed out") ||
+      err.message === "fetch failed"
+    );
+  }
+  return false;
+}
+
+export async function validateAndPersistApiKey(
+  config: Config,
+  key: string,
+  baseUrl: string,
+): Promise<void> {
+  process.stderr.write("Testing key... ");
+  const testConfig = { ...config, apiKey: key, baseUrl };
+  const requestOpts = {
+    url: chatEndpoint(testConfig.baseUrl),
+    method: "POST",
+    timeout: Math.min(config.timeout, 30),
+    body: {
+      model: "qwen3.7-max",
+      messages: [{ role: "user", content: "hi" }],
+      max_tokens: 1,
+    },
+  };
+
+  for (let attempt = 1; attempt <= 3; attempt++) {
+    try {
+      await requestJson<unknown>(testConfig, requestOpts);
+      break;
+    } catch (err) {
+      if (attempt >= 3 || !canRetry(err)) {
+        process.stderr.write("Failed\n");
+        throw new BailianError("API key validation failed", ExitCode.AUTH, "Invalid API key.", {
+          cause: err,
+        });
+      }
+      const delayMs = RETRY_DELAY_BASE_MS * 2 ** (attempt - 1);
+      await new Promise((resolve) => setTimeout(resolve, delayMs));
+    }
+  }
+
+  process.stderr.write("Valid\n");
+  const existing = readConfigFile() as Record<string, unknown>;
+  existing.api_key = key;
+  await writeConfigFile(existing);
+}
+
 export async function runConsoleLogin(
  consoleOrigin: string,
-  opts?: { needApiKey?: boolean; onApiKey?: (key: string) => Promise<void> },
+  config: Config,
+  opts?: { needApiKey?: boolean },
 ): Promise<void> {
  const state = randomBytes(16).toString("hex");
  let callbackError: unknown;
@@ -301,18 +464,35 @@ export async function runConsoleLogin(
        return;
      }

-      const { accessToken, apiKey } = await extractCredentialsFromRequest(req);
+      const {
+        accessToken,
+        apiKey,
+        baseUrl,
+        consoleSite,
+        consoleRegion,
+        consoleSwitchAgent,
+        workspaceId,
+      } = await extractCredentialsFromRequest(req);

-      if (accessToken || apiKey) {
+      const hasConfig =
+        accessToken || baseUrl || consoleSite || consoleRegion || consoleSwitchAgent || workspaceId;
+
+      if (hasConfig || apiKey) {
        try {
-          if (accessToken) {
+          if (hasConfig) {
            const existing = readConfigFile() as Record<string, unknown>;
-            existing.access_token = accessToken;
+            if (accessToken) existing.access_token = accessToken;
+            if (baseUrl) existing.base_url = baseUrl;
+            if (consoleSite) existing.console_site = consoleSite;
+            if (consoleRegion) existing.console_region = consoleRegion;
+            if (consoleSwitchAgent) existing.console_switch_agent = Number(consoleSwitchAgent);
+            if (workspaceId) existing.workspace_id = workspaceId;
            await writeConfigFile(existing);
-            process.stderr.write(`access_token saved to ${getConfigPath()}\n`);
+            process.stderr.write(`Config saved to ${getConfigPath()}\n`);
          }
-          if (apiKey && opts?.onApiKey) {
-            await opts.onApiKey(apiKey);
+          if (apiKey) {
+            const testBaseUrl = baseUrl || config.baseUrl;
+            await validateAndPersistApiKey(config, apiKey, testBaseUrl);
          }
        } catch (err: unknown) {
          callbackError = err;
@@ -329,7 +509,7 @@ export async function runConsoleLogin(
      });
      res.end("OK\n");

-      if (accessToken || apiKey) {
+      if (hasConfig || apiKey) {
        server.close();
      }
    } catch {
@@ -1,13 +1,8 @@
 import {
-  BailianError,
-  ExitCode,
-  chatEndpoint,
  defineCommand,
-  getConfigPath,
  isInteractive,
  maskToken,
  readConfigFile,
-  requestJson,
  writeConfigFile,
  type Config,
  type GlobalFlags,
@@ -16,78 +11,27 @@ import { printQuickStart } from "../../output/banner.ts";
 import { emitBare } from "../../output/output.ts";
 import { promptConfirm } from "../../output/prompt.ts";
 import { printCurrentCommandHelp } from "../../utils/command-help.ts";
-import { resolveConsoleOrigin, runConsoleLogin } from "./login-console.ts";
-
-const RETRY_DELAY_BASE_MS = 500;
-
-function canRetry(err: unknown): boolean {
-  if (err instanceof BailianError) {
-    if (err.exitCode === ExitCode.NETWORK || err.exitCode === ExitCode.TIMEOUT) {
-      return true;
-    }
-    const status = err.api?.httpStatus;
-    return status === 401 || (status !== undefined && status >= 500);
-  }
-  if (err instanceof Error) {
-    return (
-      err.name === "AbortError" ||
-      err.name === "TimeoutError" ||
-      err.message.includes("timed out") ||
-      err.message === "fetch failed"
-    );
-  }
-  return false;
-}
-
-async function validateKeyAndPersist(config: Config, key: string): Promise<void> {
-  process.stderr.write("Testing key... ");
-  const testConfig = { ...config, apiKey: key };
-  const requestOpts = {
-    url: chatEndpoint(testConfig.baseUrl),
-    method: "POST",
-    timeout: Math.min(config.timeout, 30),
-    body: {
-      model: "qwen3.7-max",
-      messages: [{ role: "user", content: "hi" }],
-      max_tokens: 1,
-    },
-  };
-
-  for (let attempt = 1; attempt <= 3; attempt++) {
-    try {
-      await requestJson<unknown>(testConfig, requestOpts);
-      break;
-    } catch (err) {
-      if (attempt >= 3 || !canRetry(err)) {
-        process.stderr.write("\n");
-        throw new BailianError("API key validation failed", ExitCode.AUTH, "Invalid API key.", {
-          cause: err,
-        });
-      }
-      // retry delay: 500ms, 1000ms, 2000ms
-      const delayMs = RETRY_DELAY_BASE_MS * 2 ** (attempt - 1);
-      await new Promise((resolve) => setTimeout(resolve, delayMs));
-    }
-  }
-
-  process.stderr.write("Valid\n");
-
-  const existing = readConfigFile() as Record<string, unknown>;
-  existing.api_key = key;
-  await writeConfigFile(existing);
-  process.stderr.write(`Saved to ${getConfigPath()}\n`);
-}
+import {
+  resolveConsoleOrigin,
+  runConsoleLogin,
+  validateAndPersistApiKey,
+} from "./login-console.ts";

 export default defineCommand({
  name: "auth login",
  description: "Authenticate with API key or console browser login (credentials can coexist)",
+  skipDefaultApiKeySetup: true,
  usage: "bl auth login --api-key <key> | bl auth login --console",
  options: [
    { flag: "--api-key <key>", description: "DashScope API key to store" },
+    {
+      flag: "--base-url <url>",
+      description: "DashScope API base URL (used with --api-key for validation)",
+    },
    {
      flag: "--console",
-      description: "Sign in via browser; opens the console login URL in your default browser",
-      type: "boolean",
+      description:
+        "Sign in via browser; use --console-site to choose domestic (default) or international",
    },
  ],
  examples: ["bl auth login --api-key sk-xxxxx", "bl auth login --console"],
@@ -100,9 +44,8 @@ export default defineCommand({
        return;
      }
      const hasApiKey = !!(config.apiKey || config.fileApiKey);
-      await runConsoleLogin(resolveConsoleOrigin(), {
+      await runConsoleLogin(resolveConsoleOrigin(config.consoleSite || "domestic"), config, {
        needApiKey: !hasApiKey,
-        onApiKey: (key) => validateKeyAndPersist(config, key),
      });
      return;
    }
@@ -130,8 +73,16 @@ export default defineCommand({
      process.exit(0);
    }

+    const baseUrl = (flags.baseUrl as string) || undefined;
+    const effectiveConfig = baseUrl ? { ...config, baseUrl } : config;
+
    if (!config.dryRun) {
-      await validateKeyAndPersist(config, key);
+      if (baseUrl) {
+        const existing = readConfigFile() as Record<string, unknown>;
+        existing.base_url = baseUrl;
+        await writeConfigFile(existing);
+      }
+      await validateAndPersistApiKey(effectiveConfig, key, effectiveConfig.baseUrl);
      printQuickStart();
    } else {
      emitBare("Would validate and save API key.");
@@ -20,6 +20,7 @@ async function clearConsoleToken(): Promise<boolean> {
 export default defineCommand({
  name: "auth logout",
  description: "Clear stored credentials",
+  skipDefaultApiKeySetup: true,
  usage: "bl auth logout [--console] [--yes] [--dry-run]",
  options: [
    {
@@ -142,6 +142,18 @@ export default defineCommand({
  name: "auth status",
  description: "Show current authentication state",
  usage: "bl auth status",
+  options: [
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
  examples: ["bl auth status", "bl auth status --output json"],
  async run(config: Config, _flags: GlobalFlags) {
    const format = detectOutputFormat(config.output);
@@ -36,9 +36,16 @@ import speechRecognize from "./speech/recognize.ts";
 import fileUpload from "./file/upload.ts";
 import consoleCall from "./console/call.ts";
 import usageFree from "./usage/free.ts";
+import usageFreetier from "./usage/freetier.ts";
+import usageStats from "./usage/stats.ts";
 import pipelineRun from "./pipeline/run.ts";
 import pipelineValidate from "./pipeline/validate.ts";
 import advisorRecommend from "./advisor/recommend.ts";
+import workspaceList from "./workspace/list.ts";
+import quotaList from "./quota/list.ts";
+import quotaRequest from "./quota/request.ts";
+import quotaHistory from "./quota/history.ts";
+import quotaCheck from "./quota/check.ts";

 /** Command registry map (no dependency on registry.ts — safe for build-time import). */
 export const commands: Record<string, Command> = {
@@ -74,11 +81,18 @@ export const commands: Record<string, Command> = {
  "file upload": fileUpload,
  "console call": consoleCall,
  "usage free": usageFree,
+  "usage freetier": usageFreetier,
+  "usage stats": usageStats,
  "pipeline run": pipelineRun,
  "pipeline validate": pipelineValidate,
  "config show": configShow,
  "config set": configSet,
  "config export-schema": configExportSchema,
  "advisor recommend": advisorRecommend,
+  "workspace list": workspaceList,
+  "quota list": quotaList,
+  "quota request": quotaRequest,
+  "quota history": quotaHistory,
+  "quota check": quotaCheck,
  update: update,
 };
@@ -13,6 +13,7 @@ export default defineCommand({
  name: "config export-schema",
  description:
    "Export all (or one) CLI command(s) as Anthropic/OpenAI-compatible JSON tool schemas",
+  skipDefaultApiKeySetup: true,
  usage: 'bl config export-schema [--command "<name>"]',
  options: [
    {
@@ -12,7 +12,6 @@ import {
 import { emitResult } from "../../output/output.ts";

 const VALID_KEYS = [
-  "region",
  "base_url",
  "output",
  "output_dir",
@@ -53,12 +52,13 @@ const KEY_ALIASES: Record<string, string> = {
 export default defineCommand({
  name: "config set",
  description: "Set a config value",
+  skipDefaultApiKeySetup: true,
  usage: "bl config set --key <key> --value <value>",
  options: [
    {
      flag: "--key <key>",
      description:
-        "Config key (region, base_url, output, output_dir, timeout, api_key, access_token, default_*_model, access_key_id, access_key_secret, workspace_id)",
+        "Config key (base_url, output, output_dir, timeout, api_key, access_token, default_*_model, access_key_id, access_key_secret, workspace_id)",
    },
    { flag: "--value <value>", description: "Value to set" },
  ],
@@ -90,13 +90,6 @@ export default defineCommand({
    }

    // Validate specific values
-    if (resolvedKey === "region" && !["cn", "us", "intl"].includes(value)) {
-      throw new BailianError(
-        `Invalid region "${value}". Valid values: cn, us, intl`,
-        ExitCode.USAGE,
-      );
-    }
-
    if (resolvedKey === "output" && !["text", "json"].includes(value)) {
      throw new BailianError(
        `Invalid output format "${value}". Valid values: text, json`,
@@ -12,6 +12,7 @@ import { emitResult } from "../../output/output.ts";
 export default defineCommand({
  name: "config show",
  description: "Display current configuration",
+  skipDefaultApiKeySetup: true,
  usage: "bl config show",
  examples: ["bl config show", "bl config show --output json"],
  async run(config: Config, _flags: GlobalFlags) {
@@ -19,25 +20,21 @@ export default defineCommand({
    const format = detectOutputFormat(config.output);

    const result: Record<string, unknown> = {
-      region: config.region,
+      ...file,
      base_url: config.baseUrl,
      output: config.output,
      timeout: config.timeout,
      config_file: getConfigPath(),
    };

-    // Mask API key if present
-    if (file.api_key) {
-      result.api_key = maskToken(file.api_key);
+    if (typeof result.api_key === "string") result.api_key = maskToken(result.api_key);
+    if (typeof result.access_token === "string")
+      result.access_token = maskToken(result.access_token);
+    if (typeof result.access_key_id === "string")
+      result.access_key_id = maskToken(result.access_key_id);
+    if (typeof result.access_key_secret === "string") {
+      result.access_key_secret = maskToken(result.access_key_secret);
    }
-    if (file.access_token) {
-      result.access_token = maskToken(file.access_token);
-    }
-
-    // Default models
-    if (file.default_text_model) result.default_text_model = file.default_text_model;
-    if (file.default_video_model) result.default_video_model = file.default_video_model;
-    if (file.default_image_model) result.default_image_model = file.default_image_model;

    emitResult(result, format);
  },
@@ -1,6 +1,7 @@
 import {
  defineCommand,
  callConsoleGateway,
+  effectiveConsoleGatewayConfig,
  resolveConsoleGatewayCredential,
  CONSOLE_GATEWAY_NO_TOKEN_MESSAGE,
  BailianError,
@@ -14,6 +15,7 @@ import { emitResult } from "../../output/output.ts";
 export default defineCommand({
  name: "console call",
  description: "Call a Bailian console API via the CLI gateway",
+  skipDefaultApiKeySetup: true,
  usage: "bl console call --api <api> --data <json> [flags]",
  options: [
    {
@@ -26,14 +28,20 @@ export default defineCommand({
      description: "Request data as JSON string",
      required: true,
    },
+    { flag: "--console-region <region>", description: "Console region" },
    {
-      flag: "--region <region>",
-      description: "API region (default: cn-beijing)",
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
    },
  ],
  examples: [
    `bl console call --api zeldaEasy.broadscope-bailian.freeTrial.queryFreeTierQuota --data '{"queryFreeTierQuotaRequest":{"models":["qwen3-max"]}}'`,
-    `bl console call --api some.api.name --data '{"key":"value"}' --region cn-beijing`,
+    `bl console call --api some.api.name --data '{"key":"value"}' --console-region cn-beijing`,
  ],
  async run(config: Config, flags: GlobalFlags) {
    const api = flags.api as string;
@@ -50,7 +58,6 @@ export default defineCommand({
      process.exit(1);
    }

-    const region = (flags.region as string) || "cn-beijing";
    const format = detectOutputFormat(config.output);

    let token: string | undefined;
@@ -63,14 +70,21 @@ export default defineCommand({
    }

    if (config.dryRun) {
-      emitResult({ api, data, region, token: token ? token.slice(0, 8) + "..." : null }, format);
+      emitResult(
+        {
+          api,
+          data,
+          token: token ? token.slice(0, 8) + "..." : null,
+          ...effectiveConsoleGatewayConfig(config),
+        },
+        format,
+      );
      return;
    }

    const result = await callConsoleGateway(config, token, {
      api,
      data,
-      region,
    });

    emitResult(result, format);
@@ -12,7 +12,6 @@ import { emitResult, emitBare } from "../../output/output.ts";
 export default defineCommand({
  name: "file upload",
  description: "Upload a local file to DashScope temporary storage (48h)",
-  apiDocs: "/developer-reference/get-temporary-file-url",
  usage: "bl file upload --file <path> --model <model>",
  options: [
    {
@@ -32,7 +32,6 @@ import {
 export default defineCommand({
  name: "image edit",
  description: "Edit an existing image with text instructions (Qwen-Image)",
-  apiDocs: "/developer-reference/qwen-image-edit-api",
  usage: "bl image edit --image <url> --prompt <text> [flags]",
  options: [
    {
@@ -65,11 +64,11 @@ export default defineCommand({
    { flag: "--out-prefix <prefix>", description: "Filename prefix (default: edited)" },
  ],
  examples: [
-    'bl image edit --image ./photo.png --prompt "把背景换成海滩"',
+    'bl image edit --image ./photo.png --prompt "Replace the background with a beach"',
    'bl image edit --image https://example.com/logo.png --prompt "Change color to blue" --n 3',
-    'bl image edit --image ./a.png --image ./b.png --prompt "把两张图合并成一张拼图"',
+    'bl image edit --image ./a.png --image ./b.png --prompt "Merge two images into one collage"',
    'bl image edit --image https://example.com/photo.png --prompt "Remove the person" --model qwen-image-2.0-pro',
-    'bl image edit --image ./photo.png --prompt "把背景换成海滩" --watermark false',
+    'bl image edit --image ./photo.png --prompt "Replace the background with a beach" --watermark false',
  ],
  async run(config: Config, flags: GlobalFlags) {
    // Normalize --image to string array (supports both single and repeated flags)
@@ -43,7 +43,6 @@ function isSyncModel(model: string): boolean {
 export default defineCommand({
  name: "image generate",
  description: "Generate images (Qwen-Image / wan2.x)",
-  apiDocs: "/best-practice/wanx/text-to-image",
  usage: "bl image generate --prompt <text> [flags]",
  options: [
    { flag: "--prompt <text>", description: "Image description", required: true },
@@ -83,7 +82,7 @@ export default defineCommand({
    },
  ],
  examples: [
-    'bl image generate --prompt "一只穿太空服的猫在火星上"',
+    'bl image generate --prompt "A cat in a spacesuit on Mars"',
    'bl image generate --prompt "Logo design" --n 3 --out-dir ./generated/',
    'bl image generate --prompt "Mountain landscape" --size 2688*1536',
    'bl image generate --prompt "A castle" --seed 42 --prompt-extend false',
@@ -1,15 +1,21 @@
 import {
  defineCommand,
+  knowledgeRetrieveEndpoint,
  signRequest,
+  requestJson,
  detectOutputFormat,
  maskToken,
+  resolveCredential,
+  trackingHeaders,
  type Config,
  type GlobalFlags,
  type KnowledgeRetrieveRequest,
  type KnowledgeRetrieveResponse,
+  type DashScopeKnowledgeRetrieveRequest,
+  type DashScopeKnowledgeRetrieveResponse,
+  type OutputFormat,
  BailianError,
  ExitCode,
-  trackingHeaders,
 } from "bailian-cli-core";
 import { failIfMissing } from "../../output/prompt.ts";
 import { emitResult, emitBare } from "../../output/output.ts";
@@ -18,24 +24,61 @@ const BAILIAN_HOST = "bailian.cn-beijing.aliyuncs.com";

 export default defineCommand({
  name: "knowledge retrieve",
-  description: "Retrieve from a Bailian knowledge base (requires AK/SK)",
+  description: "Retrieve from a Bailian knowledge base",
+  skipDefaultApiKeySetup: true,
  usage: "bl knowledge retrieve --index-id <id> --query <text> [flags]",
  options: [
    { flag: "--index-id <id>", description: "Knowledge base index ID (required)", required: true },
    { flag: "--query <text>", description: "Search query (required)", required: true },
    {
-      flag: "--workspace-id <id>",
-      description: "Bailian workspace ID (or env BAILIAN_WORKSPACE_ID)",
+      flag: "--dense-similarity-top-k <n>",
+      description: "Dense retrieval top K",
+      type: "number",
    },
-    { flag: "--top-k <n>", description: "Number of results (default: 10)", type: "number" },
-    { flag: "--rerank", description: "Enable rerank" },
+    {
+      flag: "--sparse-similarity-top-k <n>",
+      description: "Sparse retrieval top K",
+      type: "number",
+    },
+    { flag: "--rerank", description: "Enable reranking" },
    { flag: "--rerank-top-n <n>", description: "Rerank top N results", type: "number" },
-    { flag: "--access-key-id <key>", description: "Alibaba Cloud Access Key ID (or env)" },
-    { flag: "--access-key-secret <key>", description: "Alibaba Cloud Access Key Secret (or env)" },
+    {
+      flag: "--rerank-model <name>",
+      description: "Rerank model, e.g. qwen3-rerank-hybrid",
+    },
+    {
+      flag: "--rerank-mode <mode>",
+      description: "Rerank mode: qa, similar, or custom",
+    },
+    {
+      flag: "--rerank-instruct <text>",
+      description: "Custom rerank instruction, when mode=custom",
+    },
+    {
+      flag: "--top-k <n>",
+      description: "Number of results (deprecated, use --rerank-top-n)",
+      type: "number",
+    },
+    {
+      flag: "--workspace-id <id>",
+      description: "Bailian workspace ID (only needed for deprecated AK/SK auth)",
+    },
+    {
+      flag: "--access-key-id <key>",
+      description: "Deprecated: use global --api-key instead",
+    },
+    {
+      flag: "--access-key-secret <key>",
+      description: "Deprecated: use global --api-key instead",
+    },
+  ],
+  notes: [
+    "Authentication: pass `--api-key <key>`. AK/SK auth is deprecated and will be removed in a future version.",
+    "`--workspace-id` is NOT required when using --api-key.",
  ],
  examples: [
-    'bl knowledge retrieve --index-id idx_xxx --query "如何使用阿里云百炼" --workspace-id ws_xxx',
-    'bl knowledge retrieve --index-id idx_xxx --query "API限流" --top-k 5 --rerank',
+    'bl knowledge retrieve --index-id idx_xxx --query "How to use Alibaba Cloud Bailian"',
+    'bl knowledge retrieve --api-key $DASHSCOPE_API_KEY --index-id idx_xxx --query "RAG retrieval" --rerank --rerank-model qwen3-rerank-hybrid',
  ],
  async run(config: Config, flags: GlobalFlags) {
    const indexId = flags.indexId as string;
@@ -44,112 +87,226 @@ export default defineCommand({
    const query = flags.query as string;
    if (!query) failIfMissing("query", "bl knowledge retrieve --index-id <id> --query <text>");

-    const accessKeyId = (flags.accessKeyId as string) || config.accessKeyId;
-    const accessKeySecret = (flags.accessKeySecret as string) || config.accessKeySecret;
-    const workspaceId = (flags.workspaceId as string) || config.workspaceId;
-
-    if (!accessKeyId || !accessKeySecret) {
-      throw new BailianError(
-        "Knowledge retrieve requires Alibaba Cloud AK/SK.\n" +
-          "Set via: --access-key-id / --access-key-secret flags,\n" +
-          "  or env: ALIBABA_CLOUD_ACCESS_KEY_ID / ALIBABA_CLOUD_ACCESS_KEY_SECRET,\n" +
-          "  or config: bl config set access_key_id <key>",
-        ExitCode.AUTH,
-      );
-    }
-
-    if (!workspaceId) {
-      throw new BailianError(
-        "Knowledge retrieve requires a workspace ID.\n" +
-          "Set via: --workspace-id flag, or env: BAILIAN_WORKSPACE_ID, or config: bl config set workspace_id <id>",
-        ExitCode.USAGE,
-      );
-    }
-
-    const body: KnowledgeRetrieveRequest = {
-      IndexId: indexId,
-      Query: query,
-    };
-
-    if (flags.topK !== undefined) body.TopK = flags.topK as number;
-    if (flags.rerank) body.Rerank = true;
-    if (flags.rerankTopN !== undefined) body.RerankTopN = flags.rerankTopN as number;
-
    const format = detectOutputFormat(config.output);
-    const pathname = `/${workspaceId}/index/retrieve`;

-    if (config.dryRun) {
-      emitResult(
-        {
-          endpoint: `https://${BAILIAN_HOST}${pathname}`,
-          workspaceId,
-          request: body,
-        },
-        format,
-      );
-      return;
-    }
+    const hasExplicitApiKey = !!config.apiKey;
+    const hasExplicitAkSk = !!(flags.accessKeyId && flags.accessKeySecret);

-    const bodyStr = JSON.stringify(body);
-
-    const headers = signRequest({
-      accessKeyId,
-      accessKeySecret,
-      action: "Retrieve",
-      version: "2023-12-29",
-      body: bodyStr,
-      host: BAILIAN_HOST,
-      pathname,
-    });
-
-    const url = `https://${BAILIAN_HOST}${pathname}`;
-
-    if (config.verbose) {
-      process.stderr.write(`> POST ${url}\n`);
-      process.stderr.write(`> AK: ${maskToken(accessKeyId)}\n`);
-    }
-
-    const timeoutMs = config.timeout * 1000;
-    const res = await fetch(url, {
-      method: "POST",
-      headers: {
-        ...headers,
-        ...trackingHeaders(),
-      },
-      body: bodyStr,
-      signal: AbortSignal.timeout(timeoutMs),
-    });
-
-    if (config.verbose) {
-      process.stderr.write(`< ${res.status} ${res.statusText}\n`);
-    }
-
-    const data = (await res.json()) as KnowledgeRetrieveResponse & {
-      Code?: string;
-      Message?: string;
-    };
-
-    if (!res.ok || (data.Code && data.Code !== "Success")) {
-      throw new BailianError(
-        `Knowledge retrieve failed: ${data.Code || res.status} - ${data.Message || res.statusText}`,
-        ExitCode.GENERAL,
-      );
-    }
-
-    if (config.quiet || format === "text") {
-      const nodes = data.Data?.Nodes || [];
-      if (nodes.length === 0) {
-        emitBare("No results found.");
-      } else {
-        for (let i = 0; i < nodes.length; i++) {
-          const node = nodes[i];
-          emitBare(`[${i + 1}] (score: ${node.Score.toFixed(4)})`);
-          emitBare(node.Text);
-          emitBare("");
-        }
-      }
+    if (hasExplicitApiKey) {
+      await runWithApiKey(config, flags, indexId, query, format);
+    } else if (hasExplicitAkSk) {
+      await runWithAkSk(config, flags, indexId, query, format);
    } else {
-      emitResult(data, format);
+      let useApiKey = false;
+      try {
+        await resolveCredential(config);
+        useApiKey = true;
+      } catch {
+        // No API-KEY credential available
+      }
+
+      if (useApiKey) {
+        await runWithApiKey(config, flags, indexId, query, format);
+      } else {
+        await runWithAkSk(config, flags, indexId, query, format);
+      }
    }
  },
 });
+
+// ---- API-KEY path (DashScope gateway, snake_case) ----
+
+async function runWithApiKey(
+  config: Config,
+  flags: GlobalFlags,
+  indexId: string,
+  query: string,
+  format: OutputFormat,
+): Promise<void> {
+  if (flags.topK !== undefined && flags.rerankTopN === undefined) {
+    process.stderr.write("Warning: --top-k is deprecated. Use --rerank-top-n instead.\n");
+    flags.rerankTopN = flags.topK;
+  }
+
+  const body: DashScopeKnowledgeRetrieveRequest = {
+    index_id: indexId,
+    query,
+    search_filters: [],
+  };
+
+  if (flags.denseSimilarityTopK !== undefined)
+    body.dense_similarity_top_k = flags.denseSimilarityTopK as number;
+  if (flags.sparseSimilarityTopK !== undefined)
+    body.sparse_similarity_top_k = flags.sparseSimilarityTopK as number;
+  if (flags.rerank) body.enable_reranking = true;
+  if (flags.rerankTopN !== undefined) body.rerank_top_n = flags.rerankTopN as number;
+
+  if (flags.rerankModel) {
+    const rerankEntry: { model_name: string; rerank_mode?: string; rerank_instruct?: string } = {
+      model_name: flags.rerankModel as string,
+    };
+    if (flags.rerankMode) rerankEntry.rerank_mode = flags.rerankMode as string;
+    if (flags.rerankInstruct) rerankEntry.rerank_instruct = flags.rerankInstruct as string;
+    body.rerank = [rerankEntry];
+  }
+
+  const url = knowledgeRetrieveEndpoint(config.baseUrl);
+
+  if (config.dryRun) {
+    emitResult({ endpoint: url, request: body }, format);
+    return;
+  }
+
+  const response = await requestJson<DashScopeKnowledgeRetrieveResponse>(config, {
+    url,
+    method: "POST",
+    body,
+  });
+
+  const nodes = response.data?.nodes || [];
+  if (config.quiet || format === "text") {
+    emitTextNodes(nodes.map((n) => ({ text: n.text, score: n.score })));
+  } else {
+    emitResult(response, format);
+  }
+}
+
+// ---- AK/SK path (Bailian OpenAPI gateway, PascalCase) ----
+
+async function runWithAkSk(
+  config: Config,
+  flags: GlobalFlags,
+  indexId: string,
+  query: string,
+  format: OutputFormat,
+): Promise<void> {
+  const accessKeyId = (flags.accessKeyId as string) || config.accessKeyId;
+  const accessKeySecret = (flags.accessKeySecret as string) || config.accessKeySecret;
+  const workspaceId = (flags.workspaceId as string) || config.workspaceId;
+
+  if (!accessKeyId || !accessKeySecret) {
+    throw new BailianError(
+      "No credentials found.\n" +
+        "Preferred: set DASHSCOPE_API_KEY or pass --api-key.\n" +
+        "Legacy (deprecated): set ALIBABA_CLOUD_ACCESS_KEY_ID / ALIBABA_CLOUD_ACCESS_KEY_SECRET.",
+      ExitCode.AUTH,
+    );
+  }
+
+  if (!workspaceId) {
+    throw new BailianError(
+      "Knowledge retrieve requires a workspace ID.\n" +
+        "Set via: --workspace-id flag, or env: BAILIAN_WORKSPACE_ID, or config: bl config set workspace_id <id>",
+      ExitCode.USAGE,
+    );
+  }
+
+  process.stderr.write(
+    "Warning: AK/SK auth for knowledge retrieve is deprecated. Prefer --api-key or DASHSCOPE_API_KEY.\n",
+  );
+
+  const body: KnowledgeRetrieveRequest = {
+    IndexId: indexId,
+    Query: query,
+  };
+
+  if (flags.topK !== undefined && flags.rerankTopN === undefined) {
+    process.stderr.write("Warning: --top-k is deprecated. Use --rerank-top-n instead.\n");
+    flags.rerankTopN = flags.topK;
+  }
+
+  if (flags.rerank) body.EnableReranking = true;
+  if (flags.rerankTopN !== undefined) body.RerankTopN = flags.rerankTopN as number;
+  if (flags.denseSimilarityTopK !== undefined)
+    body.DenseSimilarityTopK = flags.denseSimilarityTopK as number;
+  if (flags.sparseSimilarityTopK !== undefined)
+    body.SparseSimilarityTopK = flags.sparseSimilarityTopK as number;
+
+  if (flags.rerankModel) {
+    const rerank: { ModelName: string; RerankMode?: string; RerankInstruct?: string } = {
+      ModelName: flags.rerankModel as string,
+    };
+    if (flags.rerankMode) rerank.RerankMode = flags.rerankMode as string;
+    if (flags.rerankInstruct) rerank.RerankInstruct = flags.rerankInstruct as string;
+    body.Rerank = [rerank];
+  }
+
+  const pathname = `/${workspaceId}/index/retrieve`;
+
+  if (config.dryRun) {
+    emitResult(
+      {
+        endpoint: `https://${BAILIAN_HOST}${pathname}`,
+        workspaceId,
+        request: body,
+      },
+      format,
+    );
+    return;
+  }
+
+  const bodyStr = JSON.stringify(body);
+
+  const headers = signRequest({
+    accessKeyId,
+    accessKeySecret,
+    action: "Retrieve",
+    version: "2023-12-29",
+    body: bodyStr,
+    host: BAILIAN_HOST,
+    pathname,
+  });
+
+  const url = `https://${BAILIAN_HOST}${pathname}`;
+
+  if (config.verbose) {
+    process.stderr.write(`> POST ${url}\n`);
+    process.stderr.write(`> AK: ${maskToken(accessKeyId)}\n`);
+  }
+
+  const timeoutMs = config.timeout * 1000;
+  const res = await fetch(url, {
+    method: "POST",
+    headers: { ...headers, ...trackingHeaders() },
+    body: bodyStr,
+    signal: AbortSignal.timeout(timeoutMs),
+  });
+
+  if (config.verbose) {
+    process.stderr.write(`< ${res.status} ${res.statusText}\n`);
+  }
+
+  const data = (await res.json()) as KnowledgeRetrieveResponse & {
+    Code?: string;
+    Message?: string;
+  };
+
+  if (!res.ok || (data.Code && data.Code !== "Success")) {
+    throw new BailianError(
+      `Knowledge retrieve failed: ${data.Code || res.status} - ${data.Message || res.statusText}`,
+      ExitCode.GENERAL,
+    );
+  }
+
+  const nodes = data.Data?.Nodes || [];
+  if (config.quiet || format === "text") {
+    emitTextNodes(nodes.map((n) => ({ text: n.Text, score: n.Score })));
+  } else {
+    emitResult(data, format);
+  }
+}
+
+// ---- Shared text output ----
+
+function emitTextNodes(nodes: Array<{ text: string; score: number }>): void {
+  if (nodes.length === 0) {
+    emitBare("No results found.");
+  } else {
+    for (let i = 0; i < nodes.length; i++) {
+      const node = nodes[i];
+      emitBare(`[${i + 1}] (score: ${node.score.toFixed(4)})`);
+      emitBare(node.text);
+      emitBare("");
+    }
+  }
+}
@@ -32,6 +32,7 @@ function parseArgFlags(raw: string[]): Record<string, unknown> {
 export default defineCommand({
  name: "mcp call",
  description: "Call a tool on an MCP server (tools/call)",
+  skipDefaultApiKeySetup: true,
  usage: "bl mcp call <server-code>.<tool> [--arg k=v ...] [--json '{...}'] [--url <url>]",
  options: [
    {
@@ -56,8 +57,8 @@ export default defineCommand({
    { flag: "--url <url>", description: "Override the MCP endpoint URL (for non-Bailian servers)" },
  ],
  examples: [
-    'bl mcp call market-cmapi00073529.SmartStockSelection --query "筛选ROE>15%的消费股"',
-    'bl mcp call market-cmapi00073529.FinQuery --json \'{"q":"贵州茅台","limit":5}\'',
+    'bl mcp call market-cmapi00073529.SmartStockSelection --query "Screen consumer stocks with ROE > 15%"',
+    'bl mcp call market-cmapi00073529.FinQuery --json \'{"q":"Guizhou Maotai","limit":5}\'',
    "bl mcp call market-cmapi00073529.SmartFundSelection --arg riskLevel=R3 --arg minScale=10",
  ],
  async run(config: Config, flags: GlobalFlags) {
@@ -1,6 +1,7 @@
 import {
  defineCommand,
  callConsoleGateway,
+  effectiveConsoleGatewayConfig,
  resolveConsoleGatewayCredential,
  detectOutputFormat,
  BailianError,
@@ -26,6 +27,7 @@ interface ServerSummary {
 export default defineCommand({
  name: "mcp list",
  description: "List MCP servers activated under your Bailian account",
+  skipDefaultApiKeySetup: true,
  usage: "bl mcp list [flags]",
  options: [
    { flag: "--name <text>", description: "Filter by server name (substring match)" },
@@ -35,15 +37,23 @@ export default defineCommand({
    },
    { flag: "--page <n>", description: "Page number (default: 1)", type: "number" },
    { flag: "--page-size <n>", description: "Results per page (default: 30)", type: "number" },
-    { flag: "--region <region>", description: "API region (default: cn-beijing)" },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
  ],
-  examples: ["bl mcp list", "bl mcp list --name 金融", "bl mcp list --output json"],
+  examples: ["bl mcp list", "bl mcp list --name finance", "bl mcp list --output json"],
  async run(config: Config, flags: GlobalFlags) {
    const serverName = (flags.name as string) || "";
    const type = (flags.type as string) || "OFFICIAL";
    const pageNo = (flags.page as number) || 1;
    const pageSize = (flags.pageSize as number) || 30;
-    const region = (flags.region as string) || "cn-beijing";
    const format = detectOutputFormat(config.output);

    const data = {
@@ -58,7 +68,7 @@ export default defineCommand({
    };

    if (config.dryRun) {
-      emitResult({ api: MCP_LIST_API, data, region }, format);
+      emitResult({ api: MCP_LIST_API, data, ...effectiveConsoleGatewayConfig(config) }, format);
      return;
    }

@@ -67,7 +77,6 @@ export default defineCommand({
    const result = (await callConsoleGateway(config, credential.token, {
      api: MCP_LIST_API,
      data,
-      region,
    })) as Record<string, unknown>;

    const dataField = (result?.data as Record<string, unknown> | undefined) ?? {};
@@ -13,6 +13,7 @@ import { ensureApiKey } from "../../utils/ensure-key.ts";
 export default defineCommand({
  name: "mcp tools",
  description: "List tools exposed by an MCP server (tools/list)",
+  skipDefaultApiKeySetup: true,
  usage: "bl mcp tools <server-code> [--url <url>]",
  options: [
    {
@@ -26,9 +26,9 @@ export default defineCommand({
    { flag: "--memory-library-id <id>", description: "Memory library ID (isolate memory space)" },
  ],
  examples: [
-    'bl memory add --user-id user1 --content "用户喜欢Python编程"',
-    'bl memory add --user-id user1 --messages \'[{"role":"user","content":"我喜欢旅行"}]\'',
-    'bl memory add --user-id user1 --content "住在北京" --profile-schema schema_xxx',
+    'bl memory add --user-id user1 --content "The user likes Python programming"',
+    'bl memory add --user-id user1 --messages \'[{"role":"user","content":"I like traveling"}]\'',
+    'bl memory add --user-id user1 --content "Lives in Beijing" --profile-schema schema_xxx',
  ],
  async run(config: Config, flags: GlobalFlags) {
    const userId = flags.userId as string;
@@ -20,12 +20,12 @@ export default defineCommand({
    { flag: "--description <text>", description: "Schema description" },
    {
      flag: "--attributes <json>",
-      description: 'Attributes JSON array: [{"name":"age","description":"年龄"}]',
+      description: 'Attributes JSON array: [{"name":"age","description":"age"}]',
      required: true,
    },
  ],
  examples: [
-    'bl memory profile create --name "user_basic" --attributes \'[{"name":"age","description":"年龄"},{"name":"hobby","description":"爱好"}]\'',
+    'bl memory profile create --name "user_basic" --attributes \'[{"name":"age","description":"age"},{"name":"hobby","description":"hobby"}]\'',
  ],
  async run(config: Config, flags: GlobalFlags) {
    const name = flags.name as string;
@@ -27,8 +27,8 @@ export default defineCommand({
    { flag: "--memory-library-id <id>", description: "Memory library ID" },
  ],
  examples: [
-    'bl memory search --user-id user1 --query "编程偏好"',
-    'bl memory search --user-id user1 --messages \'[{"role":"user","content":"推荐一本书"}]\' --top-k 5',
+    'bl memory search --user-id user1 --query "programming preferences"',
+    'bl memory search --user-id user1 --messages \'[{"role":"user","content":"recommend a book"}]\' --top-k 5',
  ],
  async run(config: Config, flags: GlobalFlags) {
    const userId = flags.userId as string;
@@ -24,7 +24,9 @@ export default defineCommand({
    },
    { flag: "--memory-library-id <id>", description: "Memory library ID (non-default library)" },
  ],
-  examples: ['bl memory update --node-id node_xxx --user-id user1 --content "更新后的记忆内容"'],
+  examples: [
+    'bl memory update --node-id node_xxx --user-id user1 --content "updated memory content"',
+  ],
  async run(config: Config, flags: GlobalFlags) {
    const nodeId = flags.nodeId as string;
    if (!nodeId)
@@ -1,10 +1,13 @@
 import { writeFileSync } from "fs";
+import { extname } from "path";
 import {
  defineCommand,
  request,
  chatEndpoint,
  parseSSE,
  detectOutputFormat,
+  BailianError,
+  ExitCode,
  type Config,
  type GlobalFlags,
  type ChatMessage,
@@ -13,12 +16,88 @@ import {
  type StreamChunk,
  isInteractive,
  resolveFileUrl,
+  resolveOutputDir,
+  resolveCredential,
 } from "bailian-cli-core";
 import { promptText, failIfMissing } from "../../output/prompt.ts";
 import { emitResult } from "../../output/output.ts";
-import { resolveOutputDir, resolveCredential } from "bailian-cli-core";

-const OMNI_VOICES = ["Chelsie", "Cherry", "Ethan", "Serena", "Tina"];
+interface VoiceEntry {
+  voice: string;
+  name: string;
+  desc: string;
+  lang: string;
+}
+
+// qwen-omni 系统音色
+const OMNI_VOICES: VoiceEntry[] = [
+  { voice: "Tina", name: "甜妹", desc: "甜美亲切", lang: "中文/英文" },
+  { voice: "Dylan", name: "北京-晓东", desc: "胡同少年", lang: "中文/北京" },
+  { voice: "Kiki", name: "粤语-阿清", desc: "甜美港妹", lang: "中文/英文" },
+  { voice: "Li", name: "南京-老李", desc: "南京大叔", lang: "中文/英文" },
+  { voice: "Sunny", name: "四川-晴儿", desc: "甜飒川妹", lang: "中文" },
+  { voice: "Marcus", name: "陕西-秦川", desc: "陕北汉子", lang: "中文/英文" },
+  { voice: "Eric", name: "四川-程川", desc: "成都大哥", lang: "中文/英文" },
+  { voice: "Rocky", name: "粤语-阿强", desc: "幽默港仔", lang: "中文/英文" },
+  { voice: "Jennifer", name: "詹妮弗", desc: "美剧大女主", lang: "中文/英文" },
+  { voice: "Ryan", name: "甜茶", desc: "美剧张力男", lang: "中文/英文" },
+  { voice: "Katerina", name: "卡捷琳娜", desc: "御姐深情女", lang: "中文/英文" },
+  { voice: "Peter", name: "天津-李彼得", desc: "天津捧哏", lang: "中文/英文" },
+  { voice: "Ethan", name: "晨煦", desc: "北方口音男", lang: "中文/英文" },
+];
+
+function printVoiceList(): void {
+  const col = (s: string, w: number) => s.padEnd(w);
+  process.stdout.write("\nOmni output voices:\n");
+  process.stdout.write(
+    `${col("VOICE ID", 12)} ${col("NAME", 14)} ${col("DESCRIPTION", 14)} LANGUAGE\n`,
+  );
+  process.stdout.write(`${"-".repeat(12)} ${"-".repeat(14)} ${"-".repeat(14)} ${"-".repeat(12)}\n`);
+  for (const v of OMNI_VOICES) {
+    process.stdout.write(`${col(v.voice, 12)} ${col(v.name, 14)} ${col(v.desc, 14)} ${v.lang}\n`);
+  }
+  process.stdout.write(`\nTotal: ${OMNI_VOICES.length} voices\n`);
+}
+
+/**
+ * Extension to input audio format.
+ */
+const OMNI_INPUT_AUDIO_EXT: Record<string, string> = {
+  wav: "wav",
+  mp3: "mp3",
+  amr: "amr",
+  aac: "aac",
+  m4a: "aac",
+  ogg: "ogg",
+  "3gp": "3gp",
+  "3gpp": "3gpp",
+};
+
+const audioExts = Object.keys(OMNI_INPUT_AUDIO_EXT);
+
+/**
+ * Infer the input audio format from the source URL or local file path.
+ */
+function inferInputAudioFormat(source: string): string {
+  const pathPart = source.split("?")[0].split("#")[0];
+  const ext = extname(pathPart).slice(1).toLowerCase();
+  if (!ext) {
+    throw new BailianError(
+      `Cannot infer audio format from "${source}". ` +
+        `Use a file/URL whose path ends with: ${audioExts.join(", ")}.`,
+      ExitCode.USAGE,
+    );
+  }
+  const format = OMNI_INPUT_AUDIO_EXT[ext];
+  if (!format) {
+    throw new BailianError(
+      `Unsupported audio extension ".${ext}" for "${source}". ` +
+        `Supported extensions: ${audioExts.join(", ")}.`,
+      ExitCode.USAGE,
+    );
+  }
+  return format;
+}

 /**
 * Build a standard WAV file header for PCM 16-bit mono 24kHz audio.
@@ -44,7 +123,6 @@ function buildWavHeader(dataLength: number): Buffer {
 export default defineCommand({
  name: "omni",
  description: "Multimodal chat with text + audio output (Qwen-Omni)",
-  apiDocs: "/model-studio/qwen-omni",
  usage: "bl omni --message <text> [flags]",
  options: [
    {
@@ -56,7 +134,11 @@ export default defineCommand({
    { flag: "--model <model>", description: "Model ID (default: qwen3.5-omni-plus)" },
    { flag: "--system <text>", description: "System prompt" },
    { flag: "--image <url>", description: "Image URL or local file (repeatable)", type: "array" },
-    { flag: "--audio <url>", description: "Audio URL or local file (repeatable)", type: "array" },
+    {
+      flag: "--audio <url>",
+      description: "Audio URL or local file (.wav/.mp3/.amr/.aac/.m4a/.ogg/.3gp/.3gpp)",
+      type: "array",
+    },
    {
      flag: "--video <url>",
      description: "Video file URL / local path, or comma-separated frame URLs",
@@ -64,7 +146,11 @@ export default defineCommand({
    },
    {
      flag: "--voice <voice>",
-      description: `Output voice (default: Cherry). Options: ${OMNI_VOICES.join(", ")}`,
+      description: "Output voice ID (default: Tina). Use --list-voices to see all options",
+    },
+    {
+      flag: "--list-voices",
+      description: "List available output voices and exit",
    },
    { flag: "--audio-format <fmt>", description: "Audio output format (default: wav)" },
    { flag: "--audio-out <path>", description: "Save audio to file (default: auto-generate)" },
@@ -73,16 +159,22 @@ export default defineCommand({
    { flag: "--temperature <n>", description: "Sampling temperature (0.0, 2.0]", type: "number" },
  ],
  examples: [
-    'bl omni --message "你好，你是谁？"',
-    'bl omni --message "描述这张图片" --image ./photo.jpg',
-    'bl omni --message "这段音频在说什么？" --audio https://example.com/audio.wav',
-    'bl omni --message "总结这个视频" --video https://example.com/video.mp4',
-    'bl omni --message "这个视频讲了什么" --video ./local-video.mp4 --text-only',
-    'bl omni --message "用四川话回答：今天天气怎么样" --voice Serena',
+    "bl omni --list-voices",
+    'bl omni --message "Hello, who are you?"',
+    'bl omni --message "Describe this image" --image ./photo.jpg',
+    'bl omni --message "What is this audio saying?" --audio https://example.com/audio.wav',
+    'bl omni --message "Summarize this video" --video https://example.com/video.mp4',
+    'bl omni --message "What is this video about?" --video ./local-video.mp4 --text-only',
+    'bl omni --message "Answer in Sichuan dialect: How\'s the weather today?" --voice Sunny',
    'bl omni --message "Hello" --text-only --output json',
-    'bl omni --message "朗读这段话" --audio-out greeting.wav',
+    'bl omni --message "Read this passage aloud" --audio-out greeting.wav',
  ],
  async run(config: Config, flags: GlobalFlags) {
+    if (flags.listVoices) {
+      printVoiceList();
+      return;
+    }
+
    // --- Parse messages ---
    let userMessages: string[] = [];
    if (flags.message) {
@@ -103,7 +195,7 @@ export default defineCommand({
    }

    const model = (flags.model as string) || config.defaultOmniModel || "qwen3.5-omni-plus";
-    const voice = (flags.voice as string) || "Cherry";
+    const voice = (flags.voice as string) || "Tina";
    const audioFormat = (flags.audioFormat as string) || "wav";
    const textOnly = flags.textOnly === true;
    const format = detectOutputFormat(config.output);
@@ -139,7 +231,7 @@ export default defineCommand({

    // Auto-upload local files
    const imageUrls: string[] = [];
-    const audioUrls: string[] = [];
+    const audioInputs: Array<{ source: string; data: string }> = [];
    const videoUrls: string[] = [];

    const needsResolve =
@@ -152,7 +244,7 @@ export default defineCommand({
      }
      for (const u of rawAudioUrls) {
        const resolved = await resolveFileUrl(u, credential.token, model);
-        audioUrls.push(resolved);
+        audioInputs.push({ source: u, data: resolved });
      }
      for (const u of rawVideoUrls) {
        // Detect: comma-separated = frame list, otherwise single video URL/file
@@ -174,7 +266,7 @@ export default defineCommand({
      }
    }

-    if (imageUrls.length > 0 || audioUrls.length > 0 || videoUrls.length > 0) {
+    if (imageUrls.length > 0 || audioInputs.length > 0 || videoUrls.length > 0) {
      // Find last user message and convert to multimodal content array
      for (let i = allMessages.length - 1; i >= 0; i--) {
        if (allMessages[i].role === "user") {
@@ -193,9 +285,11 @@ export default defineCommand({
            contentArray.push({ type: "image_url", image_url: { url } });
          }

-          // Add audio URLs
-          for (const url of audioUrls) {
-            contentArray.push({ type: "audio_url", audio_url: { url } });
+          for (const { source, data } of audioInputs) {
+            contentArray.push({
+              type: "input_audio",
+              input_audio: { data, format: inferInputAudioFormat(source) },
+            });
          }

          // Add video URLs: frame:xxx are frame list items, others are direct video URLs
@@ -10,6 +10,7 @@ import { loadPipelineFile } from "./load-file.ts";
 export default defineCommand({
  name: "pipeline run",
  description: "Run a pipeline workflow definition",
+  skipDefaultApiKeySetup: true,
  usage: "bl pipeline run <file> [flags]",
  options: [
    { flag: "--input <json>", description: "Runtime input as inline JSON" },
@@ -8,6 +8,7 @@ import { loadPipelineFile } from "./load-file.ts";
 export default defineCommand({
  name: "pipeline validate",
  description: "Validate a pipeline definition without executing",
+  skipDefaultApiKeySetup: true,
  usage: "bl pipeline validate <file>",
  options: [],
  examples: [
@@ -0,0 +1,341 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  effectiveConsoleGatewayConfig,
+  resolveConsoleGatewayCredential,
+  detectOutputFormat,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+import { displayWidth, padEnd } from "../../output/cjk-width.ts";
+
+const MODEL_LIST_API = "zeldaHttp.dashscopeModel./zelda/api/v1/modelCenter/listFoundationModels";
+const MONITOR_API = "zeldaEasy.bailian-telemetry.monitor.getMonitorData";
+
+interface QpmInfoItem {
+  count_limit: number;
+  count_limit_period: number;
+  usage_limit: number;
+  usage_limit_period: number;
+  usage_limit_field: string;
+  type: string;
+}
+
+interface ModelWithQpm {
+  model: string;
+  qpmInfo?: Record<string, QpmInfoItem>;
+}
+
+interface MonitorPoint {
+  value: number;
+  timestamp: number;
+}
+
+interface MonitorMetric {
+  aggMethod: string;
+  metricName: string;
+  points: MonitorPoint[];
+}
+
+function calculateRPM(item: QpmInfoItem | undefined, fallbackPeriod?: number): number {
+  if (!item) return 0;
+  const period = item.count_limit_period || fallbackPeriod;
+  if (!period) return 0;
+  return Math.floor((item.count_limit * 60) / period);
+}
+
+function calculateTPM(item: QpmInfoItem | undefined, fallbackPeriod?: number): number {
+  if (!item) return 0;
+  const period = item.usage_limit_period || fallbackPeriod;
+  if (!period) return 0;
+  return Math.floor((item.usage_limit * 60) / period);
+}
+
+function formatNumber(num: number): string {
+  return num.toLocaleString("en-US");
+}
+
+function formatRatio(usage: number, limit: number): string {
+  if (limit <= 0) return "-";
+  const pct = Math.round((usage / limit) * 100);
+  return `${formatNumber(usage)}/${formatNumber(limit)} (${pct}%)`;
+}
+
+function getStatus(usage: number, limit: number): string {
+  if (limit <= 0) return "-";
+  const pct = (usage / limit) * 100;
+  if (pct >= 100) return "Rate Limited";
+  if (pct >= 80) return "Near limit";
+  return "Normal";
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+async function fetchAllModelsWithQpm(config: Config, token: string): Promise<ModelWithQpm[]> {
+  const allModels: ModelWithQpm[] = [];
+  let pageNo = 1;
+
+  while (true) {
+    const raw = await callConsoleGateway(config, token, {
+      api: MODEL_LIST_API,
+      data: {
+        input: {
+          pageNo,
+          pageSize: 50,
+          group: false,
+          queryQpmInfo: true,
+          ignoreWorkspaceServiceSite: true,
+          supports: { selfServiceLimitIncrease: true },
+        },
+      },
+    });
+
+    const resp = extractResponseData(raw as Record<string, unknown>);
+    const list = (resp.list as ModelWithQpm[]) ?? [];
+    const total = (resp.total as number) ?? 0;
+
+    allModels.push(...list);
+    if (allModels.length >= total || list.length === 0) break;
+    pageNo++;
+  }
+
+  return allModels;
+}
+
+async function fetchMonitorData(
+  config: Config,
+  token: string,
+  modelName: string,
+  windowMinutes: number,
+): Promise<{ rpm: number; tpm: number }> {
+  const now = Date.now();
+  const startTime = now - windowMinutes * 60 * 1000;
+
+  try {
+    const raw = await callConsoleGateway(config, token, {
+      api: MONITOR_API,
+      data: {
+        reqDTO: {
+          monitorType: "Advanced",
+          metricFilters: [
+            { aggMethod: "sum_pm", metricName: "model_total_amount" },
+            { aggMethod: "sum_pm", metricName: "model_call_count" },
+          ],
+          labelFilters: {
+            resourceId: modelName,
+            resourceType: "model",
+          },
+          startTime,
+          endTime: now,
+        },
+      },
+    });
+
+    const resp = extractResponseData(raw as Record<string, unknown>);
+    const metrics = (resp.data ?? resp) as MonitorMetric[] | Record<string, unknown>;
+    if (!Array.isArray(metrics)) return { rpm: 0, tpm: 0 };
+
+    let rpm = 0;
+    let tpm = 0;
+
+    for (const metric of metrics) {
+      if (metric.aggMethod !== "sum_pm" || !metric.points?.length) continue;
+      const lastValue = metric.points[metric.points.length - 1].value ?? 0;
+      if (metric.metricName === "model_call_count") rpm = Math.round(lastValue);
+      if (metric.metricName === "model_total_amount") tpm = Math.round(lastValue);
+    }
+
+    return { rpm, tpm };
+  } catch {
+    return { rpm: -1, tpm: -1 };
+  }
+}
+
+interface CheckRow {
+  model: string;
+  rpmUsage: number;
+  rpmLimit: number;
+  tpmUsage: number;
+  tpmLimit: number;
+}
+
+function printTable(rows: CheckRow[], noColor: boolean): void {
+  const bold = noColor ? (t: string) => t : (t: string) => `\x1b[1m${t}\x1b[0m`;
+  const dim = noColor ? (t: string) => t : (t: string) => `\x1b[2m${t}\x1b[0m`;
+  const green = noColor ? (t: string) => t : (t: string) => `\x1b[32m${t}\x1b[0m`;
+  const yellow = noColor ? (t: string) => t : (t: string) => `\x1b[33m${t}\x1b[0m`;
+  const red = noColor ? (t: string) => t : (t: string) => `\x1b[31m${t}\x1b[0m`;
+
+  const headers = ["Model", "RPM Usage/Limit", "TPM Usage/Limit", "Status"];
+
+  const tableRows = rows.map((r) => {
+    const rpmStr = r.rpmUsage < 0 ? "-" : formatRatio(r.rpmUsage, r.rpmLimit);
+    const tpmStr = r.tpmUsage < 0 ? "-" : formatRatio(r.tpmUsage, r.tpmLimit);
+    const maxPct = Math.max(
+      r.rpmLimit > 0 ? (r.rpmUsage / r.rpmLimit) * 100 : 0,
+      r.tpmLimit > 0 ? (r.tpmUsage / r.tpmLimit) * 100 : 0,
+    );
+    const status =
+      r.rpmUsage < 0
+        ? "-"
+        : getStatus(Math.max(r.rpmUsage, r.tpmUsage), Math.max(r.rpmLimit, r.tpmLimit));
+    return { cells: [r.model, rpmStr, tpmStr, status], maxPct };
+  });
+
+  if (tableRows.length === 0) {
+    process.stdout.write("No models found.\n");
+    return;
+  }
+
+  const widths = headers.map((label, col) =>
+    Math.max(displayWidth(label), ...tableRows.map((r) => displayWidth(r.cells[col]))),
+  );
+
+  const headerLine = headers.map((label, col) => bold(padEnd(label, widths[col]))).join("  ");
+  const separator = widths.map((w) => dim("─".repeat(w))).join("──");
+
+  process.stdout.write(headerLine + "\n");
+  process.stdout.write(separator + "\n");
+
+  const statusCol = 3;
+  for (const r of tableRows) {
+    const cells = r.cells.map((cell, col) => {
+      if (col === statusCol) {
+        if (cell === "Rate Limited") return red(padEnd(cell, widths[col]));
+        if (cell === "Near limit") return yellow(padEnd(cell, widths[col]));
+        if (cell === "Normal") return green(padEnd(cell, widths[col]));
+      }
+      return padEnd(cell, widths[col]);
+    });
+    process.stdout.write(cells.join("  ") + "\n");
+  }
+
+  process.stdout.write(dim(`\nTotal: ${rows.length} models`) + "\n");
+}
+
+export default defineCommand({
+  name: "quota check",
+  description: "Check current usage against rate limits",
+  skipDefaultApiKeySetup: true,
+  usage: "bl quota check [--model <model>] [flags]",
+  options: [
+    {
+      flag: "--model <model>",
+      description: "Model name(s), comma-separated",
+    },
+    {
+      flag: "--period <minutes>",
+      description: "Query usage for the last N minutes (default: 2)",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: [
+    "bl quota check",
+    "bl quota check --model qwen3.6-plus",
+    "bl quota check --period 5",
+    "bl quota check --model qwen3.6-plus,qwen-turbo",
+    "bl quota check --output json",
+  ],
+  async run(config: Config, flags: GlobalFlags) {
+    const modelFlag = (flags.model as string) || undefined;
+    const rawPeriod = Number(flags.period) || 2;
+    if (rawPeriod < 1) {
+      process.stderr.write("Error: --period must be at least 1 minute.\n");
+      process.exit(1);
+    }
+    const windowMinutes = rawPeriod;
+    const format = detectOutputFormat(config.output);
+
+    if (config.dryRun) {
+      emitResult(
+        {
+          apis: [MODEL_LIST_API, MONITOR_API],
+          ...effectiveConsoleGatewayConfig(config),
+        },
+        format,
+      );
+      return;
+    }
+
+    const credential = await resolveConsoleGatewayCredential(config);
+
+    let models = await fetchAllModelsWithQpm(config, credential.token);
+
+    if (modelFlag) {
+      const names = new Set(
+        modelFlag
+          .split(",")
+          .map((n) => n.trim())
+          .filter(Boolean),
+      );
+      models = models.filter((m) => names.has(m.model));
+    }
+
+    models = models.filter((m) => m.qpmInfo);
+
+    if (models.length === 0) {
+      process.stdout.write("No models found.\n");
+      return;
+    }
+
+    const monitorResults = await Promise.all(
+      models.map((m) => fetchMonitorData(config, credential.token, m.model, windowMinutes)),
+    );
+
+    const checkRows: CheckRow[] = models.map((m, idx) => {
+      const qpm = m.qpmInfo!;
+      const modelDefault = qpm["model-default"];
+      const userSpec = qpm["user-spec"];
+
+      const rpmLimit =
+        calculateRPM(userSpec, modelDefault?.count_limit_period) || calculateRPM(modelDefault);
+      const tpmLimit =
+        calculateTPM(userSpec, modelDefault?.usage_limit_period) || calculateTPM(modelDefault);
+
+      return {
+        model: m.model,
+        rpmUsage: monitorResults[idx].rpm,
+        rpmLimit,
+        tpmUsage: monitorResults[idx].tpm,
+        tpmLimit,
+      };
+    });
+
+    if (format === "json") {
+      emitResult(checkRows, format);
+      return;
+    }
+
+    printTable(checkRows, config.noColor);
+  },
+});
@@ -0,0 +1,187 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  resolveConsoleGatewayCredential,
+  detectOutputFormat,
+  BailianError,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+import { displayWidth, padEnd } from "../../output/cjk-width.ts";
+
+const HISTORY_API = "zeldaEasy.broadscope-platform.modelInstance.listModelLimitApplications";
+
+interface LimitApplicationItem {
+  gmtCreate: string;
+  deployedModel: string;
+  usageLimit: number;
+  endTime?: string;
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+function formatDateTime(ts: string | undefined): string {
+  if (!ts) return "-";
+  try {
+    const date = new Date(ts);
+    if (isNaN(date.getTime())) return ts;
+    const y = date.getFullYear();
+    const mo = String(date.getMonth() + 1).padStart(2, "0");
+    const d = String(date.getDate()).padStart(2, "0");
+    const h = String(date.getHours()).padStart(2, "0");
+    const mi = String(date.getMinutes()).padStart(2, "0");
+    return `${y}-${mo}-${d} ${h}:${mi}`;
+  } catch {
+    return ts;
+  }
+}
+
+function formatNumber(num: number): string {
+  return num.toLocaleString("en-US");
+}
+
+function printTable(records: LimitApplicationItem[], noColor: boolean, total: number): void {
+  const bold = noColor ? (t: string) => t : (t: string) => `\x1b[1m${t}\x1b[0m`;
+  const dim = noColor ? (t: string) => t : (t: string) => `\x1b[2m${t}\x1b[0m`;
+
+  const headers = ["Model", "Token Limit", "Applied At"];
+
+  const rows = records.map((r) => [
+    r.deployedModel,
+    formatNumber(r.usageLimit),
+    formatDateTime(r.gmtCreate),
+  ]);
+
+  const widths = headers.map((label, col) =>
+    Math.max(displayWidth(label), ...rows.map((row) => displayWidth(row[col]))),
+  );
+
+  const headerLine = headers.map((label, col) => bold(padEnd(label, widths[col]))).join("  ");
+  const separator = widths.map((w) => dim("─".repeat(w))).join("──");
+
+  process.stdout.write(headerLine + "\n");
+  process.stdout.write(separator + "\n");
+
+  for (const row of rows) {
+    process.stdout.write(row.map((cell, col) => padEnd(cell, widths[col])).join("  ") + "\n");
+  }
+
+  process.stdout.write(dim(`\nTotal: ${total} records`) + "\n");
+}
+
+export default defineCommand({
+  name: "quota history",
+  description: "View quota change history",
+  skipDefaultApiKeySetup: true,
+  usage: "bl quota history [flags]",
+  options: [
+    {
+      flag: "--page <n>",
+      description: "Page number (default: 1)",
+    },
+    {
+      flag: "--page-size <n>",
+      description: "Page size (default: 10)",
+    },
+    {
+      flag: "--model <model>",
+      description: "Filter by model name",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: [
+    "bl quota history",
+    "bl quota history --page 2",
+    "bl quota history --page-size 20",
+    "bl quota history --model qwen-turbo",
+    "bl quota history --output json",
+  ],
+  async run(config: Config, flags: GlobalFlags) {
+    const page = Number(flags.page) || 1;
+    const pageSize = Number(flags.pageSize) || 10;
+    const modelFilter = (flags.model as string) || undefined;
+    const format = detectOutputFormat(config.output);
+
+    const requestData = {
+      input: { pageNo: page, pageSize },
+    };
+
+    if (config.dryRun) {
+      emitResult({ api: HISTORY_API, data: requestData }, format);
+      return;
+    }
+
+    const credential = await resolveConsoleGatewayCredential(config);
+
+    let result: unknown;
+    try {
+      result = await callConsoleGateway(config, credential.token, {
+        api: HISTORY_API,
+        data: requestData,
+      });
+    } catch (err) {
+      if (err instanceof BailianError && err.message.includes("NotLogined")) {
+        process.stderr.write(
+          "Error: session expired. Run `bl auth login --console` to re-authenticate.\n",
+        );
+        process.exit(1);
+      }
+      throw err;
+    }
+
+    const resp = extractResponseData(result as Record<string, unknown>);
+    let records = (resp.records as LimitApplicationItem[]) ?? [];
+    const total = (resp.items as number) ?? records.length;
+
+    if (modelFilter) {
+      records = records.filter((r) => r.deployedModel === modelFilter);
+    }
+
+    if (format === "json") {
+      const items = records.map((r) => ({
+        model: r.deployedModel,
+        tokenLimit: r.usageLimit,
+        appliedAt: formatDateTime(r.gmtCreate),
+      }));
+      emitResult({ records: items, total: modelFilter ? records.length : total }, format);
+      return;
+    }
+
+    if (records.length === 0) {
+      process.stdout.write("No quota change history found.\n");
+      return;
+    }
+
+    printTable(records, config.noColor, modelFilter ? records.length : total);
+  },
+});
@@ -0,0 +1,245 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  resolveConsoleGatewayCredential,
+  detectOutputFormat,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+import { displayWidth, padEnd } from "../../output/cjk-width.ts";
+
+const MODEL_LIST_API = "zeldaHttp.dashscopeModel./zelda/api/v1/modelCenter/listFoundationModels";
+
+interface QpmInfoItem {
+  count_limit: number;
+  count_limit_period: number;
+  usage_limit: number;
+  usage_limit_period: number;
+  usage_limit_field: string;
+  type: string;
+}
+
+interface ModelWithQpm {
+  model: string;
+  qpmInfo?: Record<string, QpmInfoItem>;
+}
+
+function calculateRPM(item: QpmInfoItem | undefined, fallbackPeriod?: number): number {
+  if (!item) return 0;
+  const period = item.count_limit_period || fallbackPeriod;
+  if (!period) return 0;
+  return Math.floor((item.count_limit * 60) / period);
+}
+
+function calculateTPM(item: QpmInfoItem | undefined, fallbackPeriod?: number): number {
+  if (!item) return 0;
+  const period = item.usage_limit_period || fallbackPeriod;
+  if (!period) return 0;
+  return Math.floor((item.usage_limit * 60) / period);
+}
+
+function formatNumber(num: number): string {
+  return num.toLocaleString("en-US");
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+async function fetchAllModelsWithQpm(
+  config: Config,
+  token: string,
+  onlySelfService: boolean,
+): Promise<ModelWithQpm[]> {
+  const allModels: ModelWithQpm[] = [];
+  let pageNo = 1;
+
+  while (true) {
+    const input: Record<string, unknown> = {
+      pageNo,
+      pageSize: 50,
+      group: false,
+      queryQpmInfo: true,
+      ignoreWorkspaceServiceSite: true,
+    };
+    if (onlySelfService) {
+      input.supports = { selfServiceLimitIncrease: true };
+    }
+
+    const raw = await callConsoleGateway(config, token, {
+      api: MODEL_LIST_API,
+      data: { input },
+    });
+
+    const resp = extractResponseData(raw as Record<string, unknown>);
+    const list = (resp.list as ModelWithQpm[]) ?? [];
+    const total = (resp.total as number) ?? 0;
+
+    allModels.push(...list);
+    if (allModels.length >= total || list.length === 0) break;
+    pageNo++;
+  }
+
+  return allModels;
+}
+
+function printTable(models: ModelWithQpm[], noColor: boolean): void {
+  const bold = noColor ? (t: string) => t : (t: string) => `\x1b[1m${t}\x1b[0m`;
+  const dim = noColor ? (t: string) => t : (t: string) => `\x1b[2m${t}\x1b[0m`;
+
+  const headers = ["Model", "Req/min", "Token/min", "Max TPM"];
+
+  const rows = models.map((m) => {
+    const qpm = m.qpmInfo;
+    const modelDefault = qpm?.["model-default"];
+    const userSpec = qpm?.["user-spec"];
+
+    const defaultRPM = calculateRPM(modelDefault);
+    const defaultTPM = calculateTPM(modelDefault);
+    const currentRPM = calculateRPM(userSpec, modelDefault?.count_limit_period) || defaultRPM;
+    const currentTPM = calculateTPM(userSpec, modelDefault?.usage_limit_period) || defaultTPM;
+    const maxTPM = defaultTPM * 2;
+
+    return [
+      m.model,
+      currentRPM > 0 ? formatNumber(currentRPM) : "-",
+      currentTPM > 0 ? formatNumber(currentTPM) : "-",
+      maxTPM > 0 ? formatNumber(maxTPM) : "-",
+    ];
+  });
+
+  if (rows.length === 0) {
+    process.stdout.write("No models found.\n");
+    return;
+  }
+
+  const widths = headers.map((label, col) =>
+    Math.max(displayWidth(label), ...rows.map((row) => displayWidth(row[col]))),
+  );
+
+  const headerLine = headers.map((label, col) => bold(padEnd(label, widths[col]))).join("  ");
+  const separator = widths.map((w) => dim("─".repeat(w))).join("──");
+
+  process.stdout.write(headerLine + "\n");
+  process.stdout.write(separator + "\n");
+
+  for (const row of rows) {
+    process.stdout.write(row.map((cell, col) => padEnd(cell, widths[col])).join("  ") + "\n");
+  }
+
+  process.stdout.write(dim(`\nTotal: ${models.length} models`) + "\n");
+}
+
+export default defineCommand({
+  name: "quota list",
+  description: "View model RPM/TPM rate limits",
+  skipDefaultApiKeySetup: true,
+  usage: "bl quota list [--model <model>] [flags]",
+  options: [
+    {
+      flag: "--model <model>",
+      description: "Model name(s), comma-separated",
+    },
+    {
+      flag: "--all",
+      description: "Show all models, not just self-service ones",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: [
+    "bl quota list",
+    "bl quota list --model qwen3.6-plus",
+    "bl quota list --model qwen3.6-plus,qwen-turbo",
+    "bl quota list --all",
+    "bl quota list --output json",
+  ],
+  async run(config: Config, flags: GlobalFlags) {
+    const modelFlag = (flags.model as string) || undefined;
+    const showAll = Boolean(flags.all);
+    const format = detectOutputFormat(config.output);
+
+    if (config.dryRun) {
+      const input: Record<string, unknown> = {
+        pageNo: 1,
+        pageSize: 50,
+        group: false,
+        queryQpmInfo: true,
+        ignoreWorkspaceServiceSite: true,
+      };
+      if (!showAll) input.supports = { selfServiceLimitIncrease: true };
+      emitResult({ api: MODEL_LIST_API, data: { input } }, format);
+      return;
+    }
+
+    const credential = await resolveConsoleGatewayCredential(config);
+
+    let models = await fetchAllModelsWithQpm(config, credential.token, !showAll);
+
+    if (modelFlag) {
+      const names = new Set(
+        modelFlag
+          .split(",")
+          .map((n) => n.trim())
+          .filter(Boolean),
+      );
+      models = models.filter((m) => names.has(m.model));
+      if (models.length === 0) {
+        process.stderr.write(`Error: no matching models found for "${modelFlag}".\n`);
+        process.exit(1);
+      }
+    }
+
+    if (format === "json") {
+      const items = models.map((m) => {
+        const qpm = m.qpmInfo;
+        const modelDefault = qpm?.["model-default"];
+        const userSpec = qpm?.["user-spec"];
+
+        const defaultRPM = calculateRPM(modelDefault);
+        const defaultTPM = calculateTPM(modelDefault);
+        const currentRPM = calculateRPM(userSpec, modelDefault?.count_limit_period) || defaultRPM;
+        const currentTPM = calculateTPM(userSpec, modelDefault?.usage_limit_period) || defaultTPM;
+        const maxTPM = defaultTPM * 2;
+
+        return {
+          model: m.model,
+          rpm: currentRPM > 0 ? currentRPM : null,
+          tpm: currentTPM > 0 ? currentTPM : null,
+          maxTPM: maxTPM > 0 ? maxTPM : null,
+        };
+      });
+      emitResult(items, format);
+      return;
+    }
+
+    printTable(models, config.noColor);
+  },
+});
@@ -0,0 +1,229 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  resolveConsoleGatewayCredential,
+  detectOutputFormat,
+  BailianError,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+
+const MODEL_LIST_API = "zeldaHttp.dashscopeModel./zelda/api/v1/modelCenter/listFoundationModels";
+const UPDATE_LIMITS_API = "zeldaEasy.broadscope-platform.modelInstance.updateFoundationModelLimits";
+
+interface QpmInfoItem {
+  count_limit: number;
+  count_limit_period: number;
+  usage_limit: number;
+  usage_limit_period: number;
+  usage_limit_field: string;
+  type: string;
+}
+
+function calculateTPM(item: QpmInfoItem | undefined, fallbackPeriod?: number): number {
+  if (!item) return 0;
+  const period = item.usage_limit_period || fallbackPeriod;
+  if (!period) return 0;
+  return Math.floor((item.usage_limit * 60) / period);
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+async function fetchModelQpmInfo(
+  config: Config,
+  token: string,
+  modelName: string,
+): Promise<{ model: string; qpmInfo: Record<string, QpmInfoItem> } | undefined> {
+  const raw = await callConsoleGateway(config, token, {
+    api: MODEL_LIST_API,
+    data: {
+      input: {
+        pageNo: 1,
+        pageSize: 50,
+        name: modelName,
+        group: false,
+        queryQpmInfo: true,
+        ignoreWorkspaceServiceSite: true,
+        supports: { selfServiceLimitIncrease: true },
+      },
+    },
+  });
+
+  const resp = extractResponseData(raw as Record<string, unknown>);
+  const list = (resp.list as Array<{ model: string; qpmInfo?: Record<string, QpmInfoItem> }>) ?? [];
+  return list.find((m) => m.model === modelName && m.qpmInfo) as
+    | { model: string; qpmInfo: Record<string, QpmInfoItem> }
+    | undefined;
+}
+
+export default defineCommand({
+  name: "quota request",
+  description: "Request a temporary quota increase",
+  skipDefaultApiKeySetup: true,
+  usage: "bl quota request --model <model> --tpm <value> [flags]",
+  options: [
+    {
+      flag: "--model <model>",
+      description: "Model name (required)",
+      required: true,
+    },
+    {
+      flag: "--tpm <value>",
+      description: "Target TPM value (required)",
+      required: true,
+    },
+    {
+      flag: "--yes",
+      description: "Skip downgrade confirmation",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: [
+    "bl quota request --model qwen-turbo --tpm 100000",
+    "bl quota request --model qwen3.6-plus --tpm 8000000 --yes",
+    "bl quota request --model qwen-turbo --tpm 100000 --output json",
+  ],
+  async run(config: Config, flags: GlobalFlags) {
+    const modelName = flags.model as string;
+    if (!modelName) {
+      process.stderr.write("Error: --model is required.\n");
+      process.exit(1);
+    }
+
+    const tpmValue = Number(flags.tpm);
+    if (!tpmValue || tpmValue <= 0) {
+      process.stderr.write("Error: --tpm must be a positive number.\n");
+      process.exit(1);
+    }
+
+    const autoConfirm = Boolean(flags.yes) || config.yes;
+    const format = detectOutputFormat(config.output);
+
+    if (config.dryRun) {
+      const requestData = {
+        input: {
+          model: modelName,
+          limit: { usage_limit: tpmValue },
+        },
+      };
+      emitResult({ api: UPDATE_LIMITS_API, data: requestData }, format);
+      return;
+    }
+
+    const credential = await resolveConsoleGatewayCredential(config);
+
+    const modelInfo = await fetchModelQpmInfo(config, credential.token, modelName);
+    if (!modelInfo) {
+      process.stderr.write(
+        `Error: model "${modelName}" not found or does not support self-service quota increase.\n`,
+      );
+      process.stderr.write("Hint: run `bl quota list` to view available models.\n");
+      process.exit(1);
+    }
+
+    const modelDefault = modelInfo.qpmInfo["model-default"];
+    const userSpec = modelInfo.qpmInfo["user-spec"];
+    const minLimit = calculateTPM(modelDefault);
+    const currentLimit = calculateTPM(userSpec, modelDefault?.usage_limit_period) || minLimit;
+    const maxLimit = minLimit * 2;
+
+    if (tpmValue < minLimit || tpmValue > maxLimit) {
+      process.stderr.write(
+        `Error: TPM value ${tpmValue.toLocaleString()} is out of range.\n` +
+          `  Current: ${currentLimit.toLocaleString()}\n` +
+          `  Range:   ${minLimit.toLocaleString()} ~ ${maxLimit.toLocaleString()}\n`,
+      );
+      process.exit(1);
+    }
+
+    const requestData = {
+      input: {
+        model: modelName,
+        limit: { usage_limit: tpmValue },
+        originalQpmInfo: modelInfo.qpmInfo,
+      } as Record<string, unknown>,
+    };
+
+    const submitRequest = async (confirmedDowngrade?: boolean): Promise<unknown> => {
+      if (confirmedDowngrade) {
+        requestData.input.confirmedDowngrade = true;
+      }
+      try {
+        return await callConsoleGateway(config, credential.token, {
+          api: UPDATE_LIMITS_API,
+          data: requestData,
+        });
+      } catch (err) {
+        if (err instanceof BailianError && err.message.includes("NotLogined")) {
+          process.stderr.write(
+            "Error: session expired. Run `bl auth login --console` to re-authenticate.\n",
+          );
+          process.exit(1);
+        }
+        throw err;
+      }
+    };
+
+    let result = await submitRequest();
+    const resp = extractResponseData(result as Record<string, unknown>);
+
+    if (resp.needConfirm) {
+      const confirmCode = resp.confirmCode as string;
+
+      if (confirmCode === "Refresh_Required") {
+        process.stderr.write("Error: rate limit has been updated externally. Please retry.\n");
+        process.exit(1);
+      }
+
+      if (confirmCode === "Downgrade") {
+        if (!autoConfirm) {
+          process.stderr.write(
+            `Warning: target TPM (${tpmValue.toLocaleString()}) is lower than current (${currentLimit.toLocaleString()}).\n` +
+              "Use --yes to confirm downgrade.\n",
+          );
+          process.exit(1);
+        }
+        result = await submitRequest(true);
+      }
+    }
+
+    if (format === "json") {
+      emitResult(result, format);
+      return;
+    }
+
+    process.stdout.write(
+      `Quota updated for "${modelName}": TPM ${currentLimit.toLocaleString()} → ${tpmValue.toLocaleString()}\n`,
+    );
+  },
+});
@@ -21,9 +21,9 @@ export default defineCommand({
    { flag: "--list-tools", description: "List available MCP tools and exit" },
  ],
  examples: [
-    'bl search web --query "阿里云百炼最新功能"',
+    'bl search web --query "Alibaba Cloud Bailian latest features"',
    'bl search web --query "TypeScript 5.9 new features" --count 5',
-    'bl search web --query "今日新闻"',
+    'bl search web --query "Today\'s news"',
    "bl search web --list-tools",
  ],
  async run(config: Config, flags: GlobalFlags) {
@@ -25,7 +25,6 @@ import { emitResult, emitBare } from "../../output/output.ts";
 export default defineCommand({
  name: "speech recognize",
  description: "Recognize speech from audio files (FunAudio-ASR)",
-  apiDocs: "/developer-reference/recording-file-recognition",
  usage: "bl speech recognize --url <audio-url> [flags]",
  options: [
    {
@@ -20,12 +20,14 @@ import {
  DOCS_HOSTS,
 } from "bailian-cli-core";

-const COSYVOICE_CLONE_DESIGN_DOC = `${DOCS_HOSTS.cn}/cosyvoice-clone-design-api`;
+import { VOICE_TTS_PAGE } from "../../urls.ts";
 import { downloadFile } from "../../utils/download.ts";
 import { runConcurrent, downloadParallel, getConcurrency } from "../../utils/concurrent.ts";
 import { promptText, promptSelect, failIfMissing } from "../../output/prompt.ts";
 import { emitResult, emitBare } from "../../output/output.ts";

+const COSYVOICE_CLONE_DESIGN_DOC = `${DOCS_HOSTS.cn}/cosyvoice-clone-design-api`;
+
 interface VoiceEntry {
  voice: string;
  name: string;
@@ -37,7 +39,7 @@ interface VoiceEntry {
 const COSYVOICE_V3_FLASH_VOICES: VoiceEntry[] = [
  // 社交陪伴
  { voice: "longanyang", name: "龙安洋", desc: "阳光大男孩", lang: "中文/英文" },
-  { voice: "longanhuan", name: "龙安欢", desc: "欢脱元气女", lang: "中文/英文" },
+  { voice: "longanhuan_v3", name: "龙安欢", desc: "欢脱元气女", lang: "中文/英文" },
  { voice: "longantai_v3", name: "龙安台", desc: "嗲甜台湾女", lang: "中文/英文" },
  { voice: "longhua_v3", name: "龙华", desc: "元气甜美女", lang: "中文/英文" },
  { voice: "longcheng_v3", name: "龙橙", desc: "智慧青年男", lang: "中文/英文" },
@@ -121,12 +123,14 @@ function printVoiceList(model: string): void {
  const voices = MODEL_VOICES[model];
  if (!voices) {
    process.stdout.write(`No built-in voice list available for model: ${model}\n`);
+    process.stdout.write(`Browse voices in the console: ${VOICE_TTS_PAGE}\n`);
    return;
  }
  if (voices.length === 0) {
    process.stdout.write(`Model ${model} has no system voices.\n`);
    process.stdout.write("Use clone or design voices created via the CosyVoice API.\n");
    process.stdout.write(`See: ${COSYVOICE_CLONE_DESIGN_DOC}\n`);
+    process.stdout.write(`Browse voices in the console: ${VOICE_TTS_PAGE}\n`);
    return;
  }
  const col = (s: string, w: number) => s.padEnd(w);
@@ -139,12 +143,12 @@ function printVoiceList(model: string): void {
    process.stdout.write(`${col(v.voice, 26)} ${col(v.name, 10)} ${col(v.desc, 16)} ${v.lang}\n`);
  }
  process.stdout.write(`\nTotal: ${voices.length} voices\n`);
+  process.stdout.write(`Preview and browse more voices in the console: \n${VOICE_TTS_PAGE}\n`);
 }

 export default defineCommand({
  name: "speech synthesize",
  description: "Synthesize speech from text (CosyVoice TTS)",
-  apiDocs: "/developer-reference/cosyvoice",
  usage: "bl speech synthesize --text <text> [flags]",
  options: [
    { flag: "--text <text>", description: "Text to synthesize into speech", required: true },
@@ -157,11 +161,12 @@ export default defineCommand({
    {
      flag: "--voice <voice>",
      description:
-        "Voice ID. Use --list-voices to see system voices for cosyvoice-v3-flash; for v3.5-flash provide a clone/design voice ID",
+        "Voice ID. Use --list-voices to see built-in voices for cosyvoice-v3-flash; for v3.5-flash provide a clone/design voice ID",
    },
    {
      flag: "--list-voices",
-      description: "List available system voices for the selected model and exit",
+      description:
+        "List built-in system voices for the selected model and exit (console link shown in output)",
    },
    { flag: "--format <format>", description: "Audio format: mp3, pcm, wav, opus (default: mp3)" },
    { flag: "--sample-rate <rate>", description: "Audio sample rate in Hz (e.g. 24000)" },
@@ -172,7 +177,8 @@ export default defineCommand({
    { flag: "--language <lang>", description: "Language hint (e.g. zh, en, ja, ko, fr, de)" },
    {
      flag: "--instruction <text>",
-      description: 'Natural language instruction to control speech style (e.g. "请用温柔的语调"）',
+      description:
+        'Natural language instruction to control speech style (e.g. "Use a gentle tone"）',
    },
    { flag: "--enable-ssml", description: "Enable SSML markup parsing in input text" },
    {
@@ -183,13 +189,13 @@ export default defineCommand({
  ],
  examples: [
    "bl speech synthesize --list-voices --model cosyvoice-v3-flash",
-    'bl speech synthesize --text "你好，我是千问" --voice <voice_id>',
+    'bl speech synthesize --text "Hello, I am Qwen" --voice <voice_id>',
    'bl speech synthesize --text "Hello world" --voice <voice_id> --language en',
    "bl speech synthesize --text-file script.txt --out speech.wav --voice <voice_id>",
-    'bl speech synthesize --text "今天天气真好" --voice <voice_id> --instruction "请用温柔的语调说话"',
+    'bl speech synthesize --text "Today is a good day" --voice <voice_id> --instruction "Use a gentle tone"',
    'bl speech synthesize --text "Hello" --voice <voice_id> --format wav --sample-rate 24000',
    "# Stream to audio player (macOS)",
-    'bl speech synthesize --text "你好" --voice <voice_id> --stream | afplay -',
+    'bl speech synthesize --text "Hello" --voice <voice_id> --stream | afplay -',
    "# Pipe to ffplay",
    'bl speech synthesize --text "Hello" --voice <voice_id> --stream | ffplay -nodisp -autoexit -f s16le -ar 24000 -ac 1 -',
  ],
@@ -264,7 +270,7 @@ export default defineCommand({
        const modelVoices = MODEL_VOICES[model];
        if (modelVoices && modelVoices.length > 0) {
          throw new BailianError(
-            `--voice is required.\nRun the following to see available voices:\n  bl speech synthesize --list-voices --model ${model}`,
+            `--voice is required.\nRun the following to see available voices:\n  bl speech synthesize --list-voices --model ${model}\nBrowse more voices: ${VOICE_TTS_PAGE}`,
            ExitCode.USAGE,
          );
        } else {
@@ -70,7 +70,6 @@ function parseMessages(flags: GlobalFlags): ParsedMessages {
 export default defineCommand({
  name: "text chat",
  description: "Send a chat completion (OpenAI compatible, DashScope)",
-  apiDocs: "/compatibility-of-openai-with-dashscope",
  usage: "bl text chat --message <text> [flags]",
  options: [
    { flag: "--model <model>", description: "Model ID (default: qwen3.7-max)" },
@@ -5,7 +5,8 @@ import { defineCommand, getConfigDir } from "bailian-cli-core";
 import { CLI_VERSION } from "../version.ts";
 import { NPM_PACKAGE, fetchLatestVersion } from "../utils/update-checker.ts";

-const SKILL_NAME = "bailian-cli";
+const SKILL_SOURCE = "modelstudioai/cli";
+const SKILL_INSTALL_CMD = `npx skills add ${SKILL_SOURCE} --all -g -y`;

 /** Build the install command */
 function detectInstallCommand(): { cmd: string; label: string } {
@@ -16,11 +17,13 @@ function updateAgentSkill(colors: { green: string; yellow: string; reset: string
  const { green, yellow, reset } = colors;
  process.stderr.write("\nUpdating agent skill...\n");
  try {
-    execSync(`npx skills update ${SKILL_NAME} -g -y`, { stdio: "inherit" });
+    // Reinstall (not `skills update`) into ~/.agents/skills/ and sync to all agent apps.
+    // `--all` on `skills add` means --skill '*' --agent '*' -y (Cursor, Claude Code, etc.).
+    execSync(SKILL_INSTALL_CMD, { stdio: "inherit" });
    process.stderr.write(`${green}\u2713 Agent skill updated.${reset}\n`);
  } catch {
    process.stderr.write(
-      `${yellow}Agent skill update skipped. Run manually: npx skills update ${SKILL_NAME} -g -y${reset}\n`,
+      `${yellow}Agent skill update skipped. Run manually: ${SKILL_INSTALL_CMD}${reset}\n`,
    );
  }
 }
@@ -28,6 +31,7 @@ function updateAgentSkill(colors: { green: string; yellow: string; reset: string
 export default defineCommand({
  name: "update",
  description: "Update bl to the latest version",
+  skipDefaultApiKeySetup: true,
  usage: "bl update",
  examples: ["bl update"],
  async run() {
@@ -44,6 +48,7 @@ export default defineCommand({

    if (latest && latest === CLI_VERSION) {
      process.stderr.write(`${green}\u2713 Already up to date (${CLI_VERSION}).${reset}\n`);
+      updateAgentSkill({ green, yellow, reset });
      return;
    }

@@ -2,64 +2,359 @@ import {
  defineCommand,
  callConsoleGateway,
  resolveConsoleGatewayCredential,
+  fetchModelList,
  detectOutputFormat,
  type Config,
  type GlobalFlags,
 } from "bailian-cli-core";
-import { failIfMissing } from "../../output/prompt.ts";
 import { emitResult } from "../../output/output.ts";
+import { displayWidth, padEnd } from "../../output/cjk-width.ts";

 const FREE_TIER_API = "zeldaEasy.broadscope-bailian.freeTrial.queryFreeTierQuota";
+const FREE_TIER_ONLY_STATUS_API = "zeldaEasy.broadscope-bailian.freeTrial.queryFreeTierOnlyStatus";
+
+interface FreeTierQuota {
+  model: string;
+  quotaInitTotal: number;
+  quotaTotal: number;
+  quotaValidityPeriod: number;
+  quotaStatus: string;
+}
+
+interface FreeTierOnlyStatus {
+  model: string;
+  freeTierOnly: boolean;
+}
+
+function formatNumber(num: number): string {
+  return num.toLocaleString("en-US");
+}
+
+function formatDate(ts: number): string {
+  const date = new Date(ts);
+  const year = date.getFullYear();
+  const month = String(date.getMonth() + 1).padStart(2, "0");
+  const day = String(date.getDate()).padStart(2, "0");
+  return `${year}-${month}-${day}`;
+}
+
+function formatUsage(quota: FreeTierQuota): string {
+  if (!quota.quotaInitTotal) return "-";
+  const used = quota.quotaInitTotal - quota.quotaTotal;
+  const percent = (used / quota.quotaInitTotal) * 100;
+  return `${percent.toFixed(1)}%`;
+}
+
+const CAPABILITY_TO_TYPE: Record<string, string> = {
+  Reasoning: "Text",
+  TG: "Text",
+  VU: "Text",
+  IG: "Vision",
+  VG: "Vision",
+  "Realtime-Omni": "Multimodal",
+  "Multimodal-Omni": "Multimodal",
+  ASR: "Audio",
+  TTS: "Audio",
+  "Voice-Replication": "Audio",
+  "Realtime-Text-to-Speech": "Audio",
+  "Realtime-Voice-Replication": "Audio",
+  "Realtime-ASR": "Audio",
+  "Realtime-Audio-Translate": "Audio",
+  ME: "Embedding",
+  TR: "Embedding",
+};
+
+function resolveModelType(capabilities: string[]): string {
+  for (const cap of capabilities) {
+    const type = CAPABILITY_TO_TYPE[cap];
+    if (type) return type;
+  }
+  return "-";
+}
+
+function printTable(
+  quotas: FreeTierQuota[],
+  stopMap: Map<string, boolean>,
+  typeMap: Map<string, string>,
+  noColor: boolean,
+): void {
+  const headers = ["Model", "Type", "Remaining/Total", "Usage", "Expires", "Auto-Stop"];
+
+  const rows = quotas.map((quota) => {
+    const hasQuota = quota.quotaInitTotal != null && quota.quotaTotal != null;
+    const remaining = hasQuota ? formatNumber(quota.quotaTotal) : "-";
+    const total = hasQuota ? formatNumber(quota.quotaInitTotal) : "-";
+    const stopStatus = stopMap.get(quota.model);
+    return [
+      quota.model,
+      typeMap.get(quota.model) || "-",
+      hasQuota ? `${remaining} / ${total}` : "-",
+      formatUsage(quota),
+      quota.quotaValidityPeriod ? formatDate(quota.quotaValidityPeriod) : "-",
+      quota.quotaStatus === "UNKNOWN"
+        ? "Unsupported"
+        : stopStatus === true
+          ? "ON"
+          : stopStatus === false
+            ? "OFF"
+            : "-",
+    ];
+  });
+
+  const widths = headers.map((label, col) =>
+    Math.max(displayWidth(label), ...rows.map((row) => displayWidth(row[col]))),
+  );
+
+  const dim = noColor ? (text: string) => text : (text: string) => `\x1b[2m${text}\x1b[0m`;
+  const bold = noColor ? (text: string) => text : (text: string) => `\x1b[1m${text}\x1b[0m`;
+  const green = noColor ? (text: string) => text : (text: string) => `\x1b[32m${text}\x1b[0m`;
+  const yellow = noColor ? (text: string) => text : (text: string) => `\x1b[33m${text}\x1b[0m`;
+
+  const autoStopCol = headers.length - 1;
+  const headerLine = headers.map((label, col) => bold(padEnd(label, widths[col]))).join("  ");
+  const separator = widths.map((width) => dim("─".repeat(width))).join("──");
+
+  process.stdout.write(headerLine + "\n");
+  process.stdout.write(separator + "\n");
+
+  for (const row of rows) {
+    const cells = row.map((cell, col) => {
+      if (col === autoStopCol) {
+        if (cell === "ON") return green(padEnd(cell, widths[col]));
+        if (cell === "OFF") return yellow(padEnd(cell, widths[col]));
+      }
+      return padEnd(cell, widths[col]);
+    });
+    process.stdout.write(cells.join("  ") + "\n");
+  }
+}
+
+function extractQuotas(result: unknown): FreeTierQuota[] {
+  const root = result as Record<string, unknown>;
+  const data = root.data as Record<string, unknown> | undefined;
+  if (!data) return [];
+
+  const dataV2 = data.DataV2 as Record<string, unknown> | undefined;
+  if (dataV2) {
+    const inner = dataV2.data as Record<string, unknown> | undefined;
+    const innerData = inner?.data as Record<string, unknown> | undefined;
+    return (innerData?.freeTierQuotas as FreeTierQuota[]) || [];
+  }
+
+  const direct = data.data as Record<string, unknown> | undefined;
+  return (direct?.freeTierQuotas as FreeTierQuota[]) || [];
+}
+
+function extractFreeTierOnlyStatuses(result: unknown): FreeTierOnlyStatus[] {
+  const root = result as Record<string, unknown>;
+  const data = root.data as Record<string, unknown> | undefined;
+  if (!data) return [];
+
+  const dataV2 = data.DataV2 as Record<string, unknown> | undefined;
+  if (dataV2) {
+    const inner = dataV2.data as Record<string, unknown> | undefined;
+    const innerData = inner?.data as Record<string, unknown> | undefined;
+    return (innerData?.freeTierOnlyStatuses as FreeTierOnlyStatus[]) || [];
+  }
+
+  const direct = data.data as Record<string, unknown> | undefined;
+  return (direct?.freeTierOnlyStatuses as FreeTierOnlyStatus[]) || [];
+}
+
+interface ModelInfo {
+  name: string;
+  type: string;
+}
+
+async function fetchAllModels(config: Config, token: string): Promise<ModelInfo[]> {
+  const allModels: Record<string, unknown>[] = [];
+  let page = 1;
+  while (true) {
+    const result = await fetchModelList(config, token, { pageNo: page, pageSize: 50 });
+    allModels.push(...result.models);
+    if (allModels.length >= result.total) break;
+    page++;
+  }
+  return allModels
+    .filter((item) => typeof item.model === "string" && item.model)
+    .map((item) => ({
+      name: item.model as string,
+      type: resolveModelType((item.capabilities as string[]) || []),
+    }));
+}

 export default defineCommand({
  name: "usage free",
-  description: "Query free-tier quota for a model",
-  usage: "bl usage free --model <model> [flags]",
+  description: "Query free-tier quota for models (all models if --model is omitted)",
+  skipDefaultApiKeySetup: true,
+  usage: "bl usage free [--model <model>[,model2,...]] [flags]",
  options: [
    {
      flag: "--model <model>",
-      description: "Model name to query (e.g. qwen3-max, qwen-turbo)",
-      required: true,
+      description: "Model name(s) to query, comma-separated for multiple; omit for all models",
    },
    {
-      flag: "--region <region>",
-      description: "API region (default: cn-beijing)",
+      flag: "--expiring <days>",
+      description: "Only show quotas expiring within N days",
+    },
+    {
+      flag: "--sort <field>",
+      description: "Sort by: remaining (ascending), expires (ascending)",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
    },
  ],
  examples: [
+    "bl usage free",
    "bl usage free --model qwen3-max",
+    "bl usage free --model qwen3-max,qwen-turbo",
+    "bl usage free --expiring 30",
+    "bl usage free --sort remaining",
    "bl usage free --model qwen-turbo --output json",
-    "bl usage free --model qwen3-max --region cn-beijing",
+    "bl usage free --model qwen3-max --console-region cn-beijing",
  ],
  async run(config: Config, flags: GlobalFlags) {
-    const model = flags.model as string;
-    if (!model) failIfMissing("model", "bl usage free --model <model>");
-
-    const region = (flags.region as string) || "cn-beijing";
+    const modelFlag = (flags.model as string) || undefined;
+    const expiringDays = Number(flags.expiring) || 0;
+    const VALID_SORT_FIELDS = ["remaining", "expires"] as const;
+    const sortField = (flags.sort as string) || undefined;
+    if (sortField && !VALID_SORT_FIELDS.includes(sortField as (typeof VALID_SORT_FIELDS)[number])) {
+      process.stderr.write(
+        `Error: invalid --sort value "${sortField}". Must be one of: ${VALID_SORT_FIELDS.join(", ")}\n`,
+      );
+      process.exit(1);
+    }
    const format = detectOutputFormat(config.output);

-    const credential = await resolveConsoleGatewayCredential(config);
+    let models: string[];
+    const typeMap = new Map<string, string>();

-    const data = {
-      queryFreeTierQuotaRequest: {
-        models: [model],
-      },
+    if (modelFlag) {
+      models = [
+        ...new Set(
+          modelFlag
+            .split(",")
+            .map((name) => name.trim())
+            .filter(Boolean),
+        ),
+      ];
+    } else {
+      models = [];
+    }
+
+    const requestData = {
+      queryFreeTierQuotaRequest: { models },
    };

    if (config.dryRun) {
      emitResult(
-        { api: FREE_TIER_API, data, region, token: credential.token.slice(0, 8) + "..." },
+        {
+          api: FREE_TIER_API,
+          data: requestData,
+        },
        format,
      );
      return;
    }

-    const result = await callConsoleGateway(config, credential.token, {
-      api: FREE_TIER_API,
-      data,
-      region,
-    });
+    const credential = await resolveConsoleGatewayCredential(config);

-    emitResult(result, format);
+    if (!modelFlag) {
+      const modelInfos = await fetchAllModels(config, credential.token);
+      models = modelInfos.map((info) => info.name);
+      for (const info of modelInfos) {
+        typeMap.set(info.name, info.type);
+      }
+      requestData.queryFreeTierQuotaRequest.models = models;
+    } else {
+      const searchResults = await Promise.all(
+        models.map((name) => fetchModelList(config, credential.token, { name, pageSize: 50 })),
+      );
+      for (let idx = 0; idx < models.length; idx++) {
+        const matched = searchResults[idx].models.find((item) => item.model === models[idx]);
+        if (matched) {
+          typeMap.set(models[idx], resolveModelType((matched.capabilities as string[]) || []));
+        }
+      }
+    }
+
+    const [quotaResult, stopResult] = await Promise.all([
+      callConsoleGateway(config, credential.token, {
+        api: FREE_TIER_API,
+        data: requestData,
+      }),
+      callConsoleGateway(config, credential.token, {
+        api: FREE_TIER_ONLY_STATUS_API,
+        data: { queryFreeTierOnlyStatusRequest: { models } },
+      }),
+    ]);
+
+    const allQuotas = extractQuotas(quotaResult);
+    let quotas = modelFlag
+      ? allQuotas
+      : allQuotas.filter((quota) => quota.quotaStatus === "VALID" && quota.quotaInitTotal > 0);
+
+    if (expiringDays > 0) {
+      const cutoff = Date.now() + expiringDays * 24 * 60 * 60 * 1000;
+      quotas = quotas.filter((q) => q.quotaValidityPeriod > 0 && q.quotaValidityPeriod <= cutoff);
+    }
+
+    if (sortField === "remaining") {
+      quotas.sort((a, b) => {
+        const pctA = a.quotaInitTotal ? a.quotaTotal / a.quotaInitTotal : 0;
+        const pctB = b.quotaInitTotal ? b.quotaTotal / b.quotaInitTotal : 0;
+        return pctA - pctB;
+      });
+    } else if (sortField === "expires") {
+      quotas.sort((a, b) => (a.quotaValidityPeriod ?? 0) - (b.quotaValidityPeriod ?? 0));
+    }
+
+    const stopStatuses = extractFreeTierOnlyStatuses(stopResult);
+    const stopMap = new Map(stopStatuses.map((status) => [status.model, status.freeTierOnly]));
+
+    if (format === "json") {
+      const items = quotas.map((quota) => {
+        const hasQuota = quota.quotaInitTotal != null && quota.quotaTotal != null;
+        const used = hasQuota ? quota.quotaInitTotal - quota.quotaTotal : 0;
+        const stopStatus = stopMap.get(quota.model);
+        const autoStop =
+          quota.quotaStatus === "UNKNOWN"
+            ? "unsupported"
+            : stopStatus === true
+              ? true
+              : stopStatus === false
+                ? false
+                : null;
+        return {
+          model: quota.model,
+          type: typeMap.get(quota.model) || null,
+          remaining: hasQuota ? quota.quotaTotal : null,
+          total: hasQuota ? quota.quotaInitTotal : null,
+          usagePercent:
+            hasQuota && quota.quotaInitTotal > 0
+              ? Math.round((used / quota.quotaInitTotal) * 1000) / 10
+              : null,
+          expires: quota.quotaValidityPeriod ? formatDate(quota.quotaValidityPeriod) : null,
+          autoStop,
+        };
+      });
+      emitResult(items, format);
+      return;
+    }
+
+    if (quotas.length === 0) {
+      process.stdout.write("No free-tier quota found.\n");
+      return;
+    }
+
+    printTable(quotas, stopMap, typeMap, config.noColor);
  },
 });
@@ -0,0 +1,256 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  resolveConsoleGatewayCredential,
+  fetchModelList,
+  detectOutputFormat,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+
+const ACTIVATE_API = "zeldaEasy.broadscope-bailian.freeTrial.batchActivateFreeTierOnly";
+const DEACTIVATE_API = "zeldaEasy.broadscope-bailian.freeTrial.batchDeactivateFreeTierOnly";
+const FREE_TIER_API = "zeldaEasy.broadscope-bailian.freeTrial.queryFreeTierQuota";
+const FREE_TIER_ONLY_STATUS_API = "zeldaEasy.broadscope-bailian.freeTrial.queryFreeTierOnlyStatus";
+
+interface FreeTierQuota {
+  model: string;
+  quotaTotal: number;
+  quotaInitTotal: number;
+}
+
+interface FreeTierOnlyStatus {
+  model: string;
+  freeTierOnly: boolean;
+}
+
+interface BatchResultFailure {
+  failureModelId: string;
+  errorCode: string;
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+const POLL_INTERVAL_MS = 500;
+const MAX_POLLS = 20;
+
+async function pollUntilDone(
+  config: Config,
+  token: string,
+  api: string,
+  requestKey: string,
+  models: string[],
+): Promise<unknown> {
+  let nextTaskId: string | undefined;
+
+  for (let attempt = 0; attempt < MAX_POLLS; attempt++) {
+    const requestData = {
+      [requestKey]: nextTaskId ? { taskId: nextTaskId } : { models },
+    };
+
+    const raw = await callConsoleGateway(config, token, {
+      api,
+      data: requestData,
+    });
+
+    const resp = extractResponseData(raw as Record<string, unknown>);
+    if (resp.taskId && Object.keys(resp).length === 1) {
+      nextTaskId = resp.taskId as string;
+      await new Promise((resolve) => setTimeout(resolve, POLL_INTERVAL_MS));
+      continue;
+    }
+    return raw;
+  }
+  return null;
+}
+
+async function fetchAllModelNames(config: Config, token: string): Promise<string[]> {
+  const allModels: Record<string, unknown>[] = [];
+  let page = 1;
+  while (true) {
+    const result = await fetchModelList(config, token, { pageNo: page, pageSize: 50 });
+    allModels.push(...result.models);
+    if (allModels.length >= result.total) break;
+    page++;
+  }
+  return allModels.map((item) => item.model as string).filter(Boolean);
+}
+
+export default defineCommand({
+  name: "usage freetier",
+  description:
+    "Enable or disable auto-stop for free-tier models. Enables by default; use --off to disable",
+  skipDefaultApiKeySetup: true,
+  usage: "bl usage freetier <--model <model>[,model2,...] | --all> [--off] [flags]",
+  options: [
+    {
+      flag: "--model <model>",
+      description: "Model name(s), comma-separated for multiple",
+    },
+    {
+      flag: "--all",
+      description: "Apply to all free-tier models",
+    },
+    {
+      flag: "--on",
+      description: "Enable auto-stop (default behavior)",
+    },
+    {
+      flag: "--off",
+      description: "Disable auto-stop",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: [
+    "bl usage freetier --model qwen3-max",
+    "bl usage freetier --model qwen3-max,qwen-turbo",
+    "bl usage freetier --all",
+    "bl usage freetier --on --model qwen3-max",
+    "bl usage freetier --off --model qwen3-max",
+    "bl usage freetier --off --all",
+  ],
+  async run(config: Config, flags: GlobalFlags) {
+    const modelFlag = (flags.model as string) || undefined;
+    const all = Boolean(flags.all);
+    const off = Boolean(flags.off);
+    const format = detectOutputFormat(config.output);
+
+    if (!modelFlag && !all) {
+      process.stderr.write(
+        "Error: missing required flag. Specify --model <model>[,model2,...] or --all\n",
+      );
+      process.exit(1);
+    }
+
+    let models: string[];
+    if (modelFlag) {
+      models = [
+        ...new Set(
+          modelFlag
+            .split(",")
+            .map((name) => name.trim())
+            .filter(Boolean),
+        ),
+      ];
+    } else {
+      models = [];
+    }
+
+    const api = off ? DEACTIVATE_API : ACTIVATE_API;
+    const requestKey = off
+      ? "BatchDeactivateFreeTierOnlyRequest"
+      : "BatchActivateFreeTierOnlyRequest";
+
+    if (config.dryRun) {
+      emitResult(
+        {
+          api,
+          data: { [requestKey]: { models } },
+        },
+        format,
+      );
+      return;
+    }
+
+    const credential = await resolveConsoleGatewayCredential(config);
+
+    if (!modelFlag) {
+      models = await fetchAllModelNames(config, credential.token);
+    }
+
+    if (off) {
+      const [quotaResult, stopResult] = await Promise.all([
+        callConsoleGateway(config, credential.token, {
+          api: FREE_TIER_API,
+          data: { queryFreeTierQuotaRequest: { models } },
+        }),
+        callConsoleGateway(config, credential.token, {
+          api: FREE_TIER_ONLY_STATUS_API,
+          data: { queryFreeTierOnlyStatusRequest: { models } },
+        }),
+      ]);
+
+      const quotaData = extractResponseData(quotaResult as Record<string, unknown>);
+      const quotas = (quotaData.freeTierQuotas ?? []) as FreeTierQuota[];
+      const quotaMap = new Map(quotas.map((quota) => [quota.model, quota]));
+
+      const stopData = extractResponseData(stopResult as Record<string, unknown>);
+      const stopStatuses = (stopData.freeTierOnlyStatuses ?? []) as FreeTierOnlyStatus[];
+      const stopMap = new Map(stopStatuses.map((status) => [status.model, status.freeTierOnly]));
+
+      for (const name of models) {
+        if (stopMap.get(name) === false) {
+          process.stderr.write(`Auto-stop is already disabled for "${name}".\n`);
+          continue;
+        }
+        const quota = quotaMap.get(name);
+        if (quota && quota.quotaTotal > 0 && stopMap.get(name) === true) {
+          process.stderr.write(
+            `Cannot disable auto-stop for "${name}": free-tier quota has not been fully consumed. Please disable auto-stop after the quota is exhausted.\n`,
+          );
+          continue;
+        }
+        await pollUntilDone(config, credential.token, api, requestKey, [name]);
+        process.stdout.write(`Disabled auto-stop for "${name}".\n`);
+      }
+      return;
+    }
+
+    const jsonResults: unknown[] = [];
+    for (const name of models) {
+      const result = await pollUntilDone(config, credential.token, api, requestKey, [name]);
+      if (format === "json") {
+        jsonResults.push(result);
+        continue;
+      }
+      if (result) {
+        const resultData = extractResponseData(result as Record<string, unknown>);
+        const failureModels = (resultData.failureModels as BatchResultFailure[]) ?? [];
+        if (failureModels.length > 0) {
+          process.stderr.write(
+            `Failed to enable auto-stop for "${name}" (${failureModels[0].errorCode}).\n`,
+          );
+        } else {
+          process.stdout.write(`Enabled auto-stop for "${name}".\n`);
+        }
+      } else {
+        process.stderr.write(`Warning: operation timed out for "${name}".\n`);
+      }
+    }
+    if (format === "json") {
+      emitResult(jsonResults, format);
+    }
+  },
+});
@@ -0,0 +1,468 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  resolveConsoleGatewayCredential,
+  detectOutputFormat,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+import { displayWidth, padEnd } from "../../output/cjk-width.ts";
+
+const OVERVIEW_API = "zeldaEasy.bailian-telemetry.model.getModelUsageStatistic";
+const LIST_API = "zeldaEasy.bailian-telemetry.model.listModelUsageStatisticData";
+
+interface UsageItem {
+  key: string;
+  value: number;
+  unit: string;
+}
+
+interface OverviewStatistic {
+  callCount: number;
+  modelCount: number;
+  callSuccessCount: number;
+  usages: UsageItem[];
+}
+
+interface ModelStatisticItem {
+  model: string;
+  callSuccessCount: number;
+  usages?: UsageItem[];
+  usage?: Record<string, number | undefined>;
+}
+
+interface ListStatisticResponse {
+  list: ModelStatisticItem[];
+  totalCount: number;
+  maxResults: number;
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+const POLL_INTERVAL_MS = 500;
+const MAX_POLLS = 30;
+
+async function pollTelemetryApi(
+  config: Config,
+  token: string,
+  api: string,
+  reqDTO: Record<string, unknown>,
+): Promise<unknown> {
+  let nextTaskId: string | undefined;
+
+  for (let attempt = 0; attempt < MAX_POLLS; attempt++) {
+    const requestData = nextTaskId
+      ? { reqDTO: { ...reqDTO, asyncTaskId: nextTaskId } }
+      : { reqDTO };
+
+    const raw = await callConsoleGateway(config, token, {
+      api,
+      data: requestData,
+    });
+
+    const resp = extractResponseData(raw as Record<string, unknown>);
+
+    if (resp.taskId && Object.keys(resp).length === 1) {
+      nextTaskId = resp.taskId as string;
+      await new Promise((resolve) => setTimeout(resolve, POLL_INTERVAL_MS));
+      continue;
+    }
+
+    return raw;
+  }
+  return null;
+}
+
+function resolveWorkspaceId(config: Config, flagWorkspaceId?: string): string {
+  if (flagWorkspaceId) return flagWorkspaceId;
+  if (config.workspaceId) return config.workspaceId;
+
+  process.stderr.write(
+    "Error: workspace-id is required. Set via --workspace-id, BAILIAN_WORKSPACE_ID, or `bl config set workspace_id <id>`.\n",
+  );
+  process.stderr.write("Hint: run `bl workspace list` to view available workspaces.\n");
+  process.exit(1);
+}
+
+function formatNumber(num: number): string {
+  return num.toLocaleString("en-US");
+}
+
+function formatDate(ts: number): string {
+  const date = new Date(ts);
+  const year = date.getFullYear();
+  const month = String(date.getMonth() + 1).padStart(2, "0");
+  const day = String(date.getDate()).padStart(2, "0");
+  return `${year}-${month}-${day}`;
+}
+
+function extractOverviewData(result: unknown): OverviewStatistic | undefined {
+  const resp = extractResponseData(result as Record<string, unknown>);
+  if (resp.callSuccessCount !== undefined || resp.usages !== undefined) {
+    return resp as unknown as OverviewStatistic;
+  }
+  return undefined;
+}
+
+function extractListData(result: unknown): ListStatisticResponse {
+  const resp = extractResponseData(result as Record<string, unknown>);
+  const list = (resp.list as ModelStatisticItem[]) ?? [];
+  const totalCount = (resp.totalCount as number) ?? 0;
+  const maxResults = (resp.maxResults as number) ?? 0;
+  return { list, totalCount, maxResults };
+}
+
+function resolveUsageMap(item: ModelStatisticItem): Record<string, number> {
+  const out: Record<string, number> = {};
+  if (item.usages && Array.isArray(item.usages)) {
+    for (const entry of item.usages) {
+      if (entry.key && entry.value != null) {
+        out[entry.key] = entry.value;
+      }
+    }
+  }
+  if (item.usage && typeof item.usage === "object") {
+    for (const [key, val] of Object.entries(item.usage)) {
+      if (val != null) out[key] = val;
+    }
+  }
+  return out;
+}
+
+interface UsageLabel {
+  en: string;
+  unit?: string;
+}
+
+const USAGE_KEY_LABELS: Record<string, UsageLabel> = {
+  total_token: { en: "Total Tokens", unit: "tokens" },
+  input_token: { en: "Input Tokens", unit: "tokens" },
+  output_token: { en: "Output Tokens", unit: "tokens" },
+  input_token_cache: { en: "Cached Tokens", unit: "tokens" },
+  input_token_cache_read: { en: "Cache Read", unit: "tokens" },
+  input_token_cache_creation: { en: "Cache Creation", unit: "tokens" },
+  thinking_input_token: { en: "Thinking Input", unit: "tokens" },
+  thinking_output_token: { en: "Thinking Output", unit: "tokens" },
+  text_input_token: { en: "Text Input", unit: "tokens" },
+  purein_text_output_token: { en: "Text Output", unit: "tokens" },
+  embedding_token: { en: "Embedding", unit: "tokens" },
+  image_number: { en: "Images", unit: "images" },
+  video_duration: { en: "Video Duration", unit: "sec" },
+  content_duration: { en: "Audio Duration", unit: "sec" },
+  tts_text_number: { en: "TTS Chars", unit: "chars" },
+  total_token_avg: { en: "Avg Tokens/Req" },
+};
+
+function formatLabel(label: UsageLabel): string {
+  const unitSuffix = label.unit ? ` [${label.unit}]` : "";
+  return `${label.en}${unitSuffix}`;
+}
+
+function printOverview(
+  stat: OverviewStatistic,
+  startTime: number,
+  endTime: number,
+  days: number,
+  noColor: boolean,
+): void {
+  const bold = noColor ? (text: string) => text : (text: string) => `\x1b[1m${text}\x1b[0m`;
+  const dim = noColor ? (text: string) => text : (text: string) => `\x1b[2m${text}\x1b[0m`;
+
+  process.stdout.write(
+    `${dim("Time Range Period:")} ${formatDate(startTime)} ~ ${formatDate(endTime)} ${dim(`(${days} days)`)}\n\n`,
+  );
+
+  const rows: [string, string][] = [
+    ["Models Called", formatNumber(stat.modelCount ?? 0)],
+    ["Successful Calls", formatNumber(stat.callSuccessCount ?? 0)],
+  ];
+
+  for (const usage of stat.usages ?? []) {
+    const label = USAGE_KEY_LABELS[usage.key];
+    const text = label ? formatLabel(label) : usage.key;
+    rows.push([text, formatNumber(usage.value)]);
+  }
+
+  const maxLabel = Math.max(...rows.map(([label]) => displayWidth(label)));
+  for (const [label, value] of rows) {
+    process.stdout.write(`${bold(padEnd(label, maxLabel + 2))}${value}\n`);
+  }
+}
+
+function printModelTable(
+  items: ModelStatisticItem[],
+  startTime: number,
+  endTime: number,
+  days: number,
+  noColor: boolean,
+): void {
+  const bold = noColor ? (text: string) => text : (text: string) => `\x1b[1m${text}\x1b[0m`;
+  const dim = noColor ? (text: string) => text : (text: string) => `\x1b[2m${text}\x1b[0m`;
+
+  process.stdout.write(
+    `${dim("Time Range Period:")} ${formatDate(startTime)} ~ ${formatDate(endTime)} ${dim(`(${days} days)`)}\n\n`,
+  );
+
+  if (items.length === 0) {
+    process.stdout.write("No usage data found.\n");
+    return;
+  }
+
+  const usageKeys = new Set<string>();
+  const itemUsages = items.map((item) => {
+    const usage = resolveUsageMap(item);
+    for (const key of Object.keys(usage)) usageKeys.add(key);
+    return usage;
+  });
+
+  const orderedKeys = [...usageKeys].sort((keyA, keyB) => {
+    const order = [
+      "total_token",
+      "input_token",
+      "output_token",
+      "input_token_cache",
+      "image_number",
+      "video_duration",
+      "content_duration",
+      "tts_text_number",
+    ];
+    const idxA = order.indexOf(keyA);
+    const idxB = order.indexOf(keyB);
+    return (idxA === -1 ? 999 : idxA) - (idxB === -1 ? 999 : idxB);
+  });
+
+  const headers = ["Model", "Calls", ...orderedKeys.map((key) => USAGE_KEY_LABELS[key]?.en ?? key)];
+  const rows = items.map((item, idx) => [
+    item.model,
+    formatNumber(item.callSuccessCount ?? 0),
+    ...orderedKeys.map((key) => {
+      const val = itemUsages[idx][key];
+      return val != null ? formatNumber(val) : "-";
+    }),
+  ]);
+
+  const widths = headers.map((label, col) =>
+    Math.max(displayWidth(label), ...rows.map((row) => displayWidth(row[col]))),
+  );
+
+  const headerLine = headers.map((label, col) => bold(padEnd(label, widths[col]))).join("  ");
+  const separator = widths.map((width) => dim("─".repeat(width))).join("──");
+
+  process.stdout.write(headerLine + "\n");
+  process.stdout.write(separator + "\n");
+
+  for (const row of rows) {
+    const cells = row.map((cell, col) => padEnd(cell, widths[col]));
+    process.stdout.write(cells.join("  ") + "\n");
+  }
+
+  process.stdout.write(dim(`\nTotal: ${items.length} models`) + "\n");
+}
+
+export default defineCommand({
+  name: "usage stats",
+  description: "Query model usage statistics",
+  skipDefaultApiKeySetup: true,
+  usage: "bl usage stats [--model <model>] [--days <days>] [flags]",
+  options: [
+    {
+      flag: "--model <model>",
+      description: "Model name(s), comma-separated; omit for overview",
+    },
+    {
+      flag: "--days <days>",
+      description: "Number of days (default: 7)",
+    },
+    {
+      flag: "--type <type>",
+      description: "Model type: Text, Vision, Multimodal, Audio, Embedding",
+    },
+    {
+      flag: "--workspace-id <id>",
+      description: "Workspace ID (env: BAILIAN_WORKSPACE_ID)",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: [
+    "bl usage stats",
+    "bl usage stats --days 30",
+    "bl usage stats --model qwen-turbo",
+    "bl usage stats --model qwen-turbo --days 7",
+    "bl usage stats --model qwen3.6-plus,deepseek-v4-pro",
+    "bl usage stats --type Text --days 14",
+    "bl usage stats --output json",
+  ],
+  async run(config: Config, flags: GlobalFlags) {
+    const modelFlag = (flags.model as string) || undefined;
+    const daysFlag = Number(flags.days) || 7;
+    const typeFlag = (flags.type as string) || undefined;
+    const format = detectOutputFormat(config.output);
+
+    const flagWorkspaceId = (flags.workspaceId as string) || undefined;
+    const workspaceId = resolveWorkspaceId(config, flagWorkspaceId);
+
+    const endTime = Date.now();
+    const startTime = endTime - daysFlag * 24 * 60 * 60 * 1000;
+
+    if (modelFlag) {
+      const models = [
+        ...new Set(
+          modelFlag
+            .split(",")
+            .map((name) => name.trim())
+            .filter(Boolean),
+        ),
+      ];
+
+      const baseReqDTO: Record<string, unknown> = {
+        startTime,
+        endTime,
+        modelCallSource: "Online",
+        filterWorkspaceId: workspaceId,
+        maxResults: 50,
+        skip: 0,
+        sortField: "success_count",
+        sortOrder: "DESC",
+      };
+      if (typeFlag) baseReqDTO.obsModelType = typeFlag;
+
+      if (config.dryRun) {
+        emitResult(
+          { api: LIST_API, data: { reqDTO: { ...baseReqDTO, model: models.join(",") } } },
+          format,
+        );
+        return;
+      }
+
+      const credential = await resolveConsoleGatewayCredential(config);
+
+      const results = await Promise.all(
+        models.map((model) =>
+          pollTelemetryApi(config, credential.token, LIST_API, { ...baseReqDTO, model }),
+        ),
+      );
+
+      const allItems: ModelStatisticItem[] = [];
+      for (const result of results) {
+        if (!result) continue;
+        const listData = extractListData(result);
+        allItems.push(...listData.list);
+      }
+
+      if (format === "json") {
+        const items = allItems.map((item) => {
+          const usage = resolveUsageMap(item);
+          const clean: Record<string, unknown> = {
+            model: item.model,
+            successfulCalls: item.callSuccessCount ?? 0,
+          };
+          for (const [key, val] of Object.entries(usage)) {
+            clean[key] = val;
+          }
+          return clean;
+        });
+        emitResult(
+          {
+            period: { start: formatDate(startTime), end: formatDate(endTime), days: daysFlag },
+            items,
+          },
+          format,
+        );
+        return;
+      }
+
+      printModelTable(allItems, startTime, endTime, daysFlag, config.noColor);
+    } else {
+      const reqDTO: Record<string, unknown> = {
+        startTime,
+        endTime,
+        modelCallSource: "Online",
+        filterWorkspaceId: workspaceId,
+      };
+      if (typeFlag) reqDTO.obsModelType = typeFlag;
+
+      if (config.dryRun) {
+        emitResult({ api: OVERVIEW_API, data: { reqDTO } }, format);
+        return;
+      }
+
+      const credential = await resolveConsoleGatewayCredential(config);
+
+      const result = await pollTelemetryApi(config, credential.token, OVERVIEW_API, reqDTO);
+      if (!result) {
+        process.stderr.write("Error: request timed out.\n");
+        process.exit(1);
+      }
+
+      const stat = extractOverviewData(result);
+
+      if (format === "json") {
+        if (!stat) {
+          emitResult(
+            {
+              period: { start: formatDate(startTime), end: formatDate(endTime), days: daysFlag },
+              modelsCalled: 0,
+              successfulCalls: 0,
+            },
+            format,
+          );
+          return;
+        }
+        emitResult(
+          {
+            period: { start: formatDate(startTime), end: formatDate(endTime), days: daysFlag },
+            modelsCalled: stat.modelCount ?? 0,
+            successfulCalls: stat.callSuccessCount ?? 0,
+            usages: (stat.usages ?? []).map((u) => ({
+              key: u.key,
+              value: u.value,
+              unit: u.unit,
+              label: USAGE_KEY_LABELS[u.key]?.en ?? u.key,
+            })),
+          },
+          format,
+        );
+        return;
+      }
+
+      if (!stat) {
+        process.stdout.write("No usage data found.\n");
+        return;
+      }
+
+      printOverview(stat, startTime, endTime, daysFlag, config.noColor);
+    }
+  },
+});
@@ -31,7 +31,6 @@ export default defineCommand({
  name: "video edit",
  description:
    "Edit a video with happyhorse-1.0-video-edit (style transfer, object replacement, etc.)",
-  apiDocs: "/best-practice/wanx/video-edit",
  usage: "bl video edit --video <url> --prompt <text> [flags]",
  options: [
    { flag: "--model <model>", description: "Model ID (default: happyhorse-1.0-video-edit)" },
@@ -40,7 +39,10 @@ export default defineCommand({
      description: "Input video URL or local file (mp4/mov, 2-10s)",
      required: true,
    },
-    { flag: "--prompt <text>", description: 'Edit instruction (e.g. "将画面转换为黏土风格")' },
+    {
+      flag: "--prompt <text>",
+      description: 'Edit instruction (e.g. "Convert the scene to a claymation style")',
+    },
    { flag: "--ref-image <url>", description: "Reference image URL (up to 4, comma-separated)" },
    {
      flag: "--negative-prompt <text>",
@@ -79,10 +81,10 @@ export default defineCommand({
    },
  ],
  examples: [
-    'bl video edit --video https://example.com/input.mp4 --prompt "将整个画面转换为黏土风格"',
-    'bl video edit --video https://example.com/input.mp4 --prompt "替换衣服为图片中的款式" --ref-image https://example.com/clothes.png',
+    'bl video edit --video https://example.com/input.mp4 --prompt "Convert the entire scene to claymation style"',
+    'bl video edit --video https://example.com/input.mp4 --prompt "Replace the outfit with the style shown in the image" --ref-image https://example.com/clothes.png',
    'bl video edit --video https://example.com/input.mp4 --prompt "Convert to anime style" --resolution 720P --download output.mp4',
-    'bl video edit --video https://example.com/input.mp4 --prompt "给视频里的小猫穿上衣服" --watermark false',
+    'bl video edit --video https://example.com/input.mp4 --prompt "Put clothes on the kitten in the video" --watermark false',
  ],
  async run(config: Config, flags: GlobalFlags) {
    // --- Validate video URL ---
@@ -28,28 +28,15 @@ import {
  BOOL_FLAG_WATERMARK,
 } from "../../utils/flag-descriptions.ts";

-// Normalize shorthand resolution (720P, 1080P) to pixel format for video generation models
-const RESOLUTION_SHORTCUTS: Record<string, string> = {
-  "720p": "1280*720",
-  "1080p": "1920*1080",
-  "480p": "832*480",
-};
-
-function normalizeResolution(res: string | undefined): string | undefined {
-  if (!res) return undefined;
-  return RESOLUTION_SHORTCUTS[res.toLowerCase()] || res;
-}
-
 export default defineCommand({
  name: "video generate",
  description:
-    "Generate a video from text or image (happyhorse-1.0-t2v / happyhorse-1.0-i2v / wan2.6-t2v)",
-  apiDocs: "/best-practice/wanx/text-to-video",
+    "Generate a video from text or image (happyhorse-1.1-t2v / happyhorse-1.1-i2v / wan2.6-t2v)",
  usage: "bl video generate --prompt <text> [--image <url>] [flags]",
  options: [
    {
      flag: "--model <model>",
-      description: "Model ID (default: happyhorse-1.0-t2v, or happyhorse-1.0-i2v with --image)",
+      description: "Model ID (default: happyhorse-1.1-t2v, or happyhorse-1.1-i2v with --image)",
    },
    { flag: "--prompt <text>", description: "Video description", required: true },
    { flag: "--image <url>", description: "Input image URL for image-to-video generation" },
@@ -57,8 +44,8 @@ export default defineCommand({
      flag: "--negative-prompt <text>",
      description: "Negative prompt to exclude unwanted content",
    },
-    { flag: "--resolution <res>", description: "Resolution (e.g. 1280*720, 960*960)" },
-    { flag: "--ratio <ratio>", description: "Aspect ratio (e.g. 16:9, 1:1)" },
+    { flag: "--resolution <res>", description: "Resolution: 720P or 1080P (default: 1080P)" },
+    { flag: "--ratio <ratio>", description: "Aspect ratio (e.g. 16:9, 9:16, 1:1)" },
    {
      flag: "--duration <seconds>",
      description: "Video duration in seconds (default: 5)",
@@ -86,10 +73,10 @@ export default defineCommand({
    },
  ],
  examples: [
-    'bl video generate --prompt "一个人在读书，静态镜头"',
+    'bl video generate --prompt "A person reading a book, static shot"',
    'bl video generate --prompt "Ocean waves at sunset." --download sunset.mp4',
-    'bl video generate --image https://example.com/cat.png --prompt "让画面中的猫动起来"',
-    'bl video generate --prompt "Mountain landscape" --resolution 1280*720 --duration 5',
+    'bl video generate --image https://example.com/cat.png --prompt "Make the cat in the scene move"',
+    'bl video generate --prompt "Mountain landscape" --resolution 720P --duration 5',
    'bl video generate --prompt "A cat playing with a ball" --watermark false',
  ],
  async run(config: Config, flags: GlobalFlags) {
@@ -111,7 +98,7 @@ export default defineCommand({
    const model =
      (flags.model as string) ||
      config.defaultVideoModel ||
-      ((flags.image as string) ? "happyhorse-1.0-i2v" : "happyhorse-1.0-t2v");
+      ((flags.image as string) ? "happyhorse-1.1-i2v" : "happyhorse-1.1-t2v");
    const format = detectOutputFormat(config.output);

    const imageUrl = flags.image as string | undefined;
@@ -131,13 +118,13 @@ export default defineCommand({
      input: {
        prompt: prompt!,
        negative_prompt: (flags.negativePrompt as string) || undefined,
-        // i2v models (happyhorse-1.0-i2v) require input.media with type 'first_frame'
+        // i2v models (happyhorse-1.1-i2v) require input.media with type 'first_frame'
        ...(resolvedImageUrl
          ? { media: [{ type: "first_frame" as const, url: resolvedImageUrl }] }
          : {}),
      },
      parameters: {
-        resolution: normalizeResolution(flags.resolution as string) || undefined,
+        resolution: (flags.resolution as string) || undefined,
        ratio: (flags.ratio as string) || undefined,
        duration: (flags.duration as number) || undefined,
        prompt_extend: promptExtend,
@@ -30,14 +30,13 @@ import {
 export default defineCommand({
  name: "video ref",
  description:
-    "Reference-to-video generation (happyhorse-1.0-r2v / wan2.6-r2v): multi-subject, multi-shot with voice",
-  apiDocs: "/best-practice/wanx/video-reference",
+    "Reference-to-video generation (happyhorse-1.1-r2v / wan2.6-r2v): multi-subject, multi-shot with voice",
  usage: "bl video ref --prompt <text> --image <url>... [--ref-video <url>...] [flags]",
  options: [
-    { flag: "--model <model>", description: "Model ID (default: happyhorse-1.0-r2v)" },
+    { flag: "--model <model>", description: "Model ID (default: happyhorse-1.1-r2v)" },
    {
      flag: "--prompt <text>",
-      description: "Video description with reference markers (图1, 视频1, etc.)",
+      description: "Video description with reference markers (image1, video1, etc.)",
      required: true,
    },
    {
@@ -60,11 +59,11 @@ export default defineCommand({
      description: "Voice URL for corresponding ref-video (pairs by position)",
      type: "array",
    },
-    { flag: "--resolution <res>", description: "Resolution: 720P or 1080P (default: 720P)" },
+    { flag: "--resolution <res>", description: "Resolution: 720P or 1080P (default: 1080P)" },
    { flag: "--ratio <ratio>", description: "Aspect ratio (16:9, 9:16, 1:1)" },
    {
      flag: "--duration <seconds>",
-      description: "Video duration in seconds (2-10, default: 5)",
+      description: "Video duration in seconds (default: 5)",
      type: "number",
    },
    {
@@ -89,11 +88,11 @@ export default defineCommand({
    },
  ],
  examples: [
-    'bl video ref --prompt "图1在草地上奔跑" --image person.jpg',
-    'bl video ref --prompt "视频1在弹吉他，图1走过来" --ref-video scene.mp4 --image person.jpg',
-    'bl video ref --prompt "图1说话" --image person.jpg --image-voice voice.mp3 --resolution 1080P',
-    'bl video ref --prompt "图1和图2在对话" --image a.jpg --image b.jpg --image-voice va.mp3 --image-voice vb.mp3',
-    'bl video ref --prompt "图1在喝水" --image person.jpg --watermark false',
+    'bl video ref --prompt "Image1 running on the grass" --image person.jpg',
+    'bl video ref --prompt "Video 1 plays guitar, Image 1 walks over" --ref-video scene.mp4 --image person.jpg',
+    'bl video ref --prompt "Image 1 speaks" --image person.jpg --image-voice voice.mp3 --resolution 1080P',
+    'bl video ref --prompt "Image 1 and Image 2 have a conversation" --image a.jpg --image b.jpg --image-voice va.mp3 --image-voice vb.mp3',
+    'bl video ref --prompt "Image 1 drinks water" --image person.jpg --watermark false',
  ],
  async run(config: Config, flags: GlobalFlags) {
    // --- Validate prompt ---
@@ -101,7 +100,7 @@ export default defineCommand({
    if (!prompt) {
      if (isInteractive({ nonInteractive: config.nonInteractive })) {
        const hint = await promptText({
-          message: "Enter your video prompt (use 图1, 视频1 to reference inputs):",
+          message: "Enter your video prompt (use Image1, Video1 to reference inputs):",
        });
        if (!hint) {
          process.stderr.write("Video generation cancelled.\n");
@@ -120,14 +119,14 @@ export default defineCommand({
      throw new BailianError(
        "At least one --image or --ref-video is required.",
        ExitCode.USAGE,
-        'bl video ref --prompt "描述" --image person.jpg',
+        'bl video ref --prompt "description" --image person.jpg',
      );
    }

    const imageVoices = (flags.imageVoice as string[] | undefined) || [];
    const videoVoices = (flags.videoVoice as string[] | undefined) || [];

-    const model = (flags.model as string) || "happyhorse-1.0-r2v";
+    const model = (flags.model as string) || "happyhorse-1.1-r2v";
    const format = detectOutputFormat(config.output);

    // --- Resolve file URLs (auto-upload local files) ---
@@ -72,8 +72,8 @@ export default defineCommand({
  ],
  examples: [
    "bl vision describe --image photo.jpg",
-    'bl vision describe --image https://example.com/photo.jpg --prompt "这只狗是什么品种？"',
-    'bl vision describe --video https://example.com/video.mp4 --prompt "总结视频内容"',
+    'bl vision describe --image https://example.com/photo.jpg --prompt "What breed is this dog?"',
+    'bl vision describe --video https://example.com/video.mp4 --prompt "Summarize the video content"',
    "bl vision describe --video ./local-video.mp4",
    'bl vision describe --image photo.png --prompt "Extract the text" --model qwen-vl-plus',
  ],
@@ -0,0 +1,142 @@
+import {
+  defineCommand,
+  callConsoleGateway,
+  resolveConsoleGatewayCredential,
+  detectOutputFormat,
+  type Config,
+  type GlobalFlags,
+} from "bailian-cli-core";
+import { emitResult } from "../../output/output.ts";
+import { displayWidth, padEnd } from "../../output/cjk-width.ts";
+
+const LIST_WORKSPACES_API = "zeldaEasy.bailian-dash-workspace.space.listWorkspaces";
+
+interface WorkspaceInfo {
+  workspaceId: string;
+  agentName: string;
+  defaultAgent: boolean;
+}
+
+function getNestedRecord(
+  obj: Record<string, unknown>,
+  key: string,
+): Record<string, unknown> | undefined {
+  const val = obj[key];
+  if (val && typeof val === "object" && !Array.isArray(val)) return val as Record<string, unknown>;
+  return undefined;
+}
+
+function extractResponseData(result: Record<string, unknown>): Record<string, unknown> {
+  const data = getNestedRecord(result, "data");
+  if (!data) return result;
+
+  const dataV2 = getNestedRecord(data, "DataV2");
+  if (dataV2) {
+    const inner = getNestedRecord(dataV2, "data");
+    const innerData = inner ? getNestedRecord(inner, "data") : undefined;
+    return innerData ?? inner ?? dataV2;
+  }
+
+  const direct = getNestedRecord(data, "data");
+  return direct ?? data;
+}
+
+function printTable(workspaces: WorkspaceInfo[], noColor: boolean): void {
+  const bold = noColor ? (text: string) => text : (text: string) => `\x1b[1m${text}\x1b[0m`;
+  const dim = noColor ? (text: string) => text : (text: string) => `\x1b[2m${text}\x1b[0m`;
+  const green = noColor ? (text: string) => text : (text: string) => `\x1b[32m${text}\x1b[0m`;
+
+  const headers = ["Name", "Workspace ID", "Default"];
+
+  const rows = workspaces.map((ws) => [
+    ws.agentName,
+    ws.workspaceId,
+    ws.defaultAgent ? "Yes" : "-",
+  ]);
+
+  const widths = headers.map((label, col) =>
+    Math.max(displayWidth(label), ...rows.map((row) => displayWidth(row[col]))),
+  );
+
+  const headerLine = headers.map((label, col) => bold(padEnd(label, widths[col]))).join("  ");
+  const separator = widths.map((width) => dim("─".repeat(width))).join("──");
+
+  process.stdout.write(headerLine + "\n");
+  process.stdout.write(separator + "\n");
+
+  for (const row of rows) {
+    const cells = row.map((cell, col) => {
+      if (col === 2 && cell === "Yes") return green(padEnd(cell, widths[col]));
+      return padEnd(cell, widths[col]);
+    });
+    process.stdout.write(cells.join("  ") + "\n");
+  }
+
+  process.stdout.write(dim(`\nTotal: ${workspaces.length}`) + "\n");
+}
+
+export default defineCommand({
+  name: "workspace list",
+  description: "List all workspaces",
+  skipDefaultApiKeySetup: true,
+  usage: "bl workspace list [flags]",
+  options: [
+    {
+      flag: "--list <n>",
+      description: "Limit number of results",
+    },
+    { flag: "--console-region <region>", description: "Console region" },
+    {
+      flag: "--console-site <site>",
+      description: "Console site: domestic, international",
+    },
+    {
+      flag: "--console-switch-agent <uid>",
+      description: "Switch agent UID",
+      type: "number",
+    },
+  ],
+  examples: ["bl workspace list", "bl workspace list --list 5", "bl workspace list --output json"],
+  async run(config: Config, flags: GlobalFlags) {
+    const limit = Number(flags.list) || 0;
+    const format = detectOutputFormat(config.output);
+
+    const credential = await resolveConsoleGatewayCredential(config);
+
+    if (config.dryRun) {
+      emitResult({ api: LIST_WORKSPACES_API, data: {} }, format);
+      return;
+    }
+
+    const result = await callConsoleGateway(config, credential.token, {
+      api: LIST_WORKSPACES_API,
+      data: {},
+    });
+
+    const resp = extractResponseData(result as Record<string, unknown>);
+    const dataArr = resp.data as Record<string, unknown>[] | undefined;
+    if (!Array.isArray(dataArr) || dataArr.length === 0) {
+      if (format === "json") {
+        emitResult([], format);
+      } else {
+        process.stdout.write("No workspace found.\n");
+      }
+      return;
+    }
+
+    let workspaces = dataArr as unknown as WorkspaceInfo[];
+    if (limit > 0) workspaces = workspaces.slice(0, limit);
+
+    if (format === "json") {
+      const items = workspaces.map((ws) => ({
+        workspaceId: ws.workspaceId,
+        name: ws.agentName,
+        default: ws.defaultAgent,
+      }));
+      emitResult(items, format);
+      return;
+    }
+
+    printTable(workspaces, config.noColor);
+  },
+});
@@ -9,6 +9,10 @@ import { API_KEY_PAGE } from "./urls.ts";

 const LABEL_WIDTH = 13;

+/** Short reminder; full resolution order matches `loadConfig` in bailian-cli-core. */
+const BASE_URL_HINT =
+  "If the DashScope host is wrong, check baseUrl (--base-url, bl config show, or DASHSCOPE_BASE_URL).";
+
 function pad(label: string): string {
  return label.padEnd(LABEL_WIDTH);
 }
@@ -22,7 +26,10 @@ function alignContinuation(text: string): string {

 function enhanceHint(err: BailianError): string | undefined {
  if (err.exitCode === ExitCode.AUTH) {
-    if (err.message === CONSOLE_GATEWAY_NO_TOKEN_MESSAGE) {
+    if (
+      err.message === CONSOLE_GATEWAY_NO_TOKEN_MESSAGE ||
+      err.hint?.includes("auth login --console")
+    ) {
      return err.hint;
    }
    return [
@@ -76,19 +83,23 @@ function pickNetworkHint(code: string | undefined): string {
  switch (code) {
    case "ENOTFOUND":
    case "EAI_AGAIN":
-      return "DNS resolution failed. Check DASHSCOPE_BASE_URL or your DNS / network.";
+      return `DNS resolution failed. Check DNS / network. ${BASE_URL_HINT}`;
    case "ECONNREFUSED":
      return "Connection refused. Check the target host/port and proxy settings.";
    case "ECONNRESET":
-      return "Connection reset by peer. Retry, or check proxy / firewall.";
+      return (
+        "Connection reset by peer. Retry, or check proxy / firewall.\n" +
+        "If you are behind a VPN or corporate proxy, route bl through it:\n" +
+        "export HTTPS_PROXY=http://127.0.0.1:<proxy-port>"
+      );
    case "ETIMEDOUT":
-      return "Connection timed out. Check your network or try a different region.";
+      return `Connection timed out. Check your network. ${BASE_URL_HINT}`;
    case "CERT_HAS_EXPIRED":
    case "UNABLE_TO_VERIFY_LEAF_SIGNATURE":
    case "DEPTH_ZERO_SELF_SIGNED_CERT":
      return "TLS certificate error. Check system clock and CA bundle.";
    default:
-      return "Check network connection, proxy settings (HTTP_PROXY / HTTPS_PROXY), and DASHSCOPE_BASE_URL.";
+      return `Check network and proxy (HTTP_PROXY / HTTPS_PROXY). ${BASE_URL_HINT}`;
  }
 }

@@ -162,9 +173,8 @@ export function handleError(err: unknown): never {
        "Request timed out.",
        ExitCode.TIMEOUT,
        "Try increasing --timeout (e.g. --timeout 60).\n" +
-          "If this happens on every request with a valid API key, you may be hitting the wrong region.\n" +
-          "Run: bl auth status   — to check your credentials and region.\n" +
-          "Run: bl config set --key region --value cn   — to override the region.",
+          `${BASE_URL_HINT}\n` +
+          "Run: bl auth status — to check credentials.",
        { cause: err },
      );
      return handleError(timeout);
@@ -3,13 +3,12 @@ import { registry } from "./registry.ts";
 import {
  GLOBAL_OPTIONS,
  loadConfig,
-  readConfigFile,
  resolveCredential,
  trackCommandExecution,
  flushTelemetry,
-  type Region,
 } from "bailian-cli-core";
 import { ensureApiKey } from "./utils/ensure-key.ts";
+import { setupProxyFromEnv } from "./proxy.ts";
 import { handleError } from "./error-handler.ts";
 import { checkForUpdate, getPendingUpdateNotification } from "./utils/update-checker.ts";
 import { maybeShowStatusBar } from "./output/status-bar.ts";
@@ -21,14 +20,15 @@ import {
  setExecutingCommandPath,
 } from "./utils/command-help.ts";

+// 必须在任何 fetch 发起前安装（含 update-checker / telemetry）
+try {
+  setupProxyFromEnv();
+} catch (err) {
+  handleError(err);
+}
+
 registerCommandHelpPrinter((commandPath, out) => {
-  const a = process.argv.slice(2);
-  const ri = a.indexOf("--region");
-  const region = ((ri >= 0 && a[ri + 1]) ||
-    process.env.DASHSCOPE_REGION ||
-    readConfigFile().region ||
-    "cn") as Region;
-  registry.printHelp(commandPath, out, region);
+  registry.printHelp(commandPath, out);
 });

 // 优雅处理 Ctrl+C
@@ -45,26 +45,6 @@ process.stdout.on("error", (e: NodeJS.ErrnoException) => {
  else throw e;
 });

-// 自己接管鉴权 或 根本不需要 API key 的命令
-const NO_AUTH_SETUP = [
-  ["auth", "login"],
-  ["auth", "logout"],
-  ["config", "show"],
-  ["config", "set"],
-  ["config", "export-schema"],
-  ["update"],
-  ["knowledge", "retrieve"],
-  ["pipeline", "run"],
-  ["pipeline", "validate"],
-  ["model", "list"],
-  ["app", "list"],
-  ["console", "call"],
-  ["usage", "free"],
-  ["mcp", "list"],
-  ["mcp", "tools"],
-  ["mcp", "call"],
-];
-
 async function main() {
  let argv = process.argv.slice(2);
  if (argv[0] === "--") argv = argv.slice(1);
@@ -77,12 +57,7 @@ async function main() {
  const commandPath = scanCommandPath(argv, GLOBAL_OPTIONS);

  if (argv.includes("--help") || argv.includes("-h")) {
-    const ri = argv.indexOf("--region");
-    const region = ((ri >= 0 && argv[ri + 1]) ||
-      process.env.DASHSCOPE_REGION ||
-      readConfigFile().region ||
-      "cn") as Region;
-    registry.printHelp(commandPath, process.stderr, region);
+    registry.printHelp(commandPath, process.stderr);
    process.exit(0);
  }

@@ -108,12 +83,7 @@ async function main() {

  // 组路径（例如 `bl speech` 未接子命令）：展示帮助后干净退出
  if (registry.isGroupPath(commandPath)) {
-    const ri = argv.indexOf("--region");
-    const region = ((ri >= 0 && argv[ri + 1]) ||
-      process.env.DASHSCOPE_REGION ||
-      readConfigFile().region ||
-      "cn") as Region;
-    registry.printHelp(commandPath, process.stderr, region);
+    registry.printHelp(commandPath, process.stderr);
    process.exit(0);
  }

@@ -126,8 +96,8 @@ async function main() {
  config.clientName = "bailian-cli";
  config.clientVersion = CLI_VERSION;

-  const needsAuthSetup = !NO_AUTH_SETUP.some((cmd) => cmd.every((c, i) => commandPath[i] === c));
-  if (needsAuthSetup) {
+  // 默认执行 ensureApiKey；自行处理鉴权或仅需 Console/AK-SK 等的命令在 defineCommand 上设 skipDefaultApiKeySetup
+  if (!command.skipDefaultApiKeySetup) {
    await ensureApiKey(config);
    try {
      const credential = await resolveCredential(config);
@@ -1,10 +1,10 @@
 import { API_KEY_PAGE } from "../urls.ts";

 const QUICK_START_TASKS = [
-  "帮我生成一套鸭舌帽的亚马逊电商主图(白底 + 场景图 + 模特上身图)",
-  "帮我生成一段 3 分钟的幽默相声音频",
-  "帮我生成一套小红帽故事绘本 PDF(含插图)",
-  "帮我分析这个视频的内容并写一篇小红书文案",
+  "Help me generate a set of Amazon e-commerce main images for baseball caps (white background + lifestyle shots + model wear shots)",
+  "Help me generate a 3-minute humorous crosstalk audio clip",
+  "Help me generate a Little Red Riding Hood picture-book PDF (with illustrations)",
+  "Help me analyze this video and write a Xiaohongshu-style post",
 ];

 function colors() {
@@ -0,0 +1,24 @@
+function isCjk(code: number): boolean {
+  return (
+    (code >= 0x2e80 && code <= 0x9fff) ||
+    (code >= 0xf900 && code <= 0xfaff) ||
+    (code >= 0xfe30 && code <= 0xfe4f) ||
+    (code >= 0xff00 && code <= 0xff60) ||
+    (code >= 0xffe0 && code <= 0xffe6) ||
+    (code >= 0x20000 && code <= 0x2fa1f)
+  );
+}
+
+export function displayWidth(text: string): number {
+  let width = 0;
+  for (const char of text) {
+    const code = char.codePointAt(0) ?? 0;
+    width += isCjk(code) ? 2 : 1;
+  }
+  return width;
+}
+
+export function padEnd(text: string, targetWidth: number): string {
+  const gap = targetWidth - displayWidth(text);
+  return gap > 0 ? text + " ".repeat(gap) : text;
+}
@@ -5,7 +5,6 @@ const reset = "\x1b[0m";
 const dim = "\x1b[2m";
 const bold = "\x1b[1m";
 const mmBlue = "\x1b[38;2;43;82;255m";
-const mmCyan = "\x1b[38;2;6;184;212m";
 const mmPink = "\x1b[38;2;236;72;153m";

 function tildePath(p: string): string {
@@ -20,7 +19,6 @@ export function maybeShowStatusBar(
  if (config.quiet || !process.stderr.isTTY) return;

  const filePath = config.configPath ? tildePath(config.configPath) : "~/.bailian/config.json";
-  const regionSrc = config.fileRegion ? `${config.fileRegion} (file)` : "cn (default)";
  const authTag = resolved
    ? `${resolved.source} · ${resolved.method}`
    : config.apiKey
@@ -32,8 +30,6 @@ export function maybeShowStatusBar(
    `${bold}${mmBlue}BAILIAN${reset} ` +
      `${dim}${filePath}${reset} ` +
      `${dim}|${reset} ` +
-      `${dim}Region:${reset} ${mmCyan}${regionSrc}${reset} ` +
-      `${dim}|${reset} ` +
      `${dim}Auth:${reset} ${mmPink}${maskedKey}${reset} ${dim}${authTag}${reset}\n`,
  );
 }
@@ -391,7 +391,7 @@ export async function videoGenerate(
    });
  }

-  const model = input.model || (input.image ? "happyhorse-1.0-i2v" : "happyhorse-1.0-t2v");
+  const model = input.model || (input.image ? "happyhorse-1.1-i2v" : "happyhorse-1.1-t2v");

  let resolvedImageUrl: string | undefined;
  if (input.image) {
@@ -0,0 +1,48 @@
+import { setGlobalDispatcher, EnvHttpProxyAgent } from "undici";
+import { BailianError, ExitCode } from "bailian-cli-core";
+
+export interface ProxyEnv {
+  httpProxy?: string;
+  httpsProxy?: string;
+  noProxy?: string;
+}
+
+function pick(env: NodeJS.ProcessEnv, ...keys: string[]): string | undefined {
+  for (const key of keys) {
+    const value = env[key]?.trim();
+    if (value) return value;
+  }
+  return undefined;
+}
+
+/**
+ * 读取代理环境变量（小写优先，与 curl 约定一致）。
+ * 空白值视为未设置——undici 自身用 `??` 取值，空字符串的小写变量会屏蔽
+ * 已设置的大写变量，这里统一清洗后显式传入，绕开该坑。
+ */
+export function readProxyEnv(env: NodeJS.ProcessEnv = process.env): ProxyEnv {
+  return {
+    httpProxy: pick(env, "http_proxy", "HTTP_PROXY"),
+    httpsProxy: pick(env, "https_proxy", "HTTPS_PROXY"),
+    noProxy: pick(env, "no_proxy", "NO_PROXY"),
+  };
+}
+
+// Node 内置 fetch（undici）默认不读取代理环境变量，VPN / 公司代理环境下会
+// 绕过代理直连而被拦截（见 issue #35）。仅当用户显式设置了 HTTP_PROXY /
+// HTTPS_PROXY 时才安装代理 dispatcher（同时支持 NO_PROXY），未设置时不触碰
+// 全局 dispatcher，行为与之前完全一致。
+export function setupProxyFromEnv(): void {
+  const { httpProxy, httpsProxy, noProxy } = readProxyEnv();
+  if (!httpProxy && !httpsProxy) return;
+  try {
+    setGlobalDispatcher(new EnvHttpProxyAgent({ httpProxy, httpsProxy, noProxy }));
+  } catch (err) {
+    throw new BailianError(
+      `Invalid proxy configuration: ${err instanceof Error ? err.message : String(err)}`,
+      ExitCode.USAGE,
+      "Check HTTP_PROXY / HTTPS_PROXY values, e.g. export HTTPS_PROXY=http://127.0.0.1:7890",
+      { cause: err },
+    );
+  }
+}
@@ -1,7 +1,7 @@
 import type { Command } from "bailian-cli-core";
 import { BailianError } from "bailian-cli-core";
 import { ExitCode } from "bailian-cli-core";
-import { DOCS_HOSTS, GLOBAL_OPTIONS, type Region } from "bailian-cli-core";
+import { GLOBAL_OPTIONS } from "bailian-cli-core";
 import { commands } from "./commands/catalog.ts";

 export type { Command, OptionDef } from "bailian-cli-core";
@@ -133,11 +133,7 @@ class CommandRegistry {
    out.isTTY ? `\x1b[38;2;59;130;246m${s}\x1b[0m` : s;
  private dim = (s: string, out: NodeJS.WriteStream) => (out.isTTY ? `\x1b[2m${s}\x1b[0m` : s);

-  printHelp(
-    commandPath: string[],
-    out: NodeJS.WriteStream = process.stdout,
-    region: Region = "cn",
-  ): void {
+  printHelp(commandPath: string[], out: NodeJS.WriteStream = process.stdout): void {
    if (commandPath.length === 0) {
      this.printRootHelp(out);
      return;
@@ -154,7 +150,7 @@ class CommandRegistry {
    }

    if (node.command) {
-      this.printCommandHelp(node.command, out, region);
+      this.printCommandHelp(node.command, out);
      return;
    }

@@ -233,7 +229,7 @@ ${b("Getting Help:")}
 `);
  }

-  private printCommandHelp(cmd: Command, out: NodeJS.WriteStream, region: Region = "cn"): void {
+  private printCommandHelp(cmd: Command, out: NodeJS.WriteStream): void {
    const b = (s: string) => this.bold(s, out);
    const a = (s: string) => this.accent(s, out);
    const d = (s: string) => this.dim(s, out);
@@ -247,15 +243,18 @@ ${b("Getting Help:")}
        out.write(`  ${a(opt.flag.padEnd(maxLen + 2))} ${d(opt.description)}\n`);
      }
    }
+    if (cmd.notes && cmd.notes.length > 0) {
+      out.write(`\n${b("Notes:")}\n`);
+      for (const note of cmd.notes) {
+        out.write(`  ${note}\n`);
+      }
+    }
    if (cmd.examples && cmd.examples.length > 0) {
      out.write(`\n${b("Examples:")}\n`);
      for (const ex of cmd.examples) {
        out.write(`  ${d(ex)}\n`);
      }
    }
-    if (cmd.apiDocs) {
-      out.write(`\n${b("API Reference:")} ${d(DOCS_HOSTS[region] + cmd.apiDocs)}\n`);
-    }
    out.write(
      `\n${d("Global flags (--api-key, --output, --quiet, etc.) are always available.")}\n`,
    );
@@ -14,3 +14,6 @@ export const BAILIAN_CONSOLE = `${BAILIAN_CONSOLE_ROOT}/cn-beijing`;

 /** Direct deep link to API key management page. */
 export const API_KEY_PAGE = `${BAILIAN_CONSOLE}/?tab=app#/api-key`;
+
+/** Voice TTS experience center — browse system and custom voices. */
+export const VOICE_TTS_PAGE = "https://help.aliyun.com/zh/model-studio/cosyvoice-voice-list";
@@ -2,21 +2,21 @@ import { describe, expect, test } from "vite-plus/test";
 import { isDashScopeE2EReady, parseStdoutJson, runCli } from "./helpers.ts";

 describe("e2e: advisor recommend", () => {
-  test("advisor 分组展示子命令帮助且成功退出", async () => {
+  test("advisor shows subcommand groups and exits successfully", async () => {
    const { stdout, stderr, exitCode } = await runCli(["advisor"]);
    expect(exitCode, stderr).toBe(0);
    expect(`${stdout}\n${stderr}`).toMatch(/advisor|recommend/i);
  });

-  test("advisor recommend --help 正常退出", async () => {
+  test("advisor recommend --help exits successfully", async () => {
    const { stderr, exitCode } = await runCli(["advisor", "recommend", "--help"]);
    expect(exitCode, stderr).toBe(0);
    expect(stderr).toMatch(/recommend|--message|dry-run/i);
  });
 });

-describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）", () => {
-  test("advisor recommend 缺少 --message 时打印帮助并退出 (0)", async () => {
+describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend (DashScope)", () => {
+  test("advisor recommend without --message prints help and exits", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
@@ -26,13 +26,13 @@ describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）",
    expect(`${stdout}\n${stderr}`).toMatch(/--message|Usage:/i);
  });

-  test("advisor recommend --dry-run 输出意图分析和候选列表", async () => {
+  test("advisor recommend --dry-run outputs intent analysis and candidates", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
      "--dry-run",
      "--message",
-      "我想做一个能理解图片的客服机器人",
+      "I want to build a customer service bot that understands images",
      "--non-interactive",
      "--output",
      "json",
@@ -44,7 +44,7 @@ describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）",
      candidateCount?: number;
      candidates?: Array<{ model?: string; score?: number }>;
    }>(stdout);
-    expect(data.userInput).toBe("我想做一个能理解图片的客服机器人");
+    expect(data.userInput).toBe("I want to build a customer service bot that understands images");
    expect(data.intent?.requiredCapabilities).toContain("VU");
    expect(data.intent?.inputModality).toContain("Image");
    expect(data.candidateCount).toBeGreaterThan(0);
@@ -52,40 +52,44 @@ describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）",
    expect(data.candidates?.[0]?.score).toBeGreaterThan(0);
  }, 60_000);

-  test("advisor recommend 完整推荐流程返回结果", async () => {
+  test("advisor recommend full flow returns results", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
      "--message",
-      "低成本高并发的在线客服",
+      "low-cost high-concurrency online customer service",
      "--non-interactive",
      "--output",
      "json",
    ]);
    expect(exitCode, stderr).toBe(0);
    const data = parseStdoutJson<{
-      type?: string;
-      recommendations?: Array<{
-        model?: string;
-        name?: string;
-        reason?: string;
-      }>;
+      intent?: { taskSummary?: string };
+      result?: {
+        type?: string;
+        recommendations?: Array<{
+          model?: string;
+          name?: string;
+          reason?: string;
+        }>;
+      };
+      candidates?: number;
    }>(stdout);
-    expect(data.type).toBe("single");
-    expect(data.recommendations?.length).toBeGreaterThan(0);
-    expect(data.recommendations?.[0]?.model).toBeDefined();
-    expect(data.recommendations?.[0]?.reason).toBeDefined();
+    expect(data.result?.type).toBe("single");
+    expect(data.result?.recommendations?.length).toBeGreaterThan(0);
+    expect(data.result?.recommendations?.[0]?.model).toBeDefined();
+    expect(data.result?.recommendations?.[0]?.reason).toBeDefined();
  }, 120_000);

-  // ---- 模型偏好：正例 ----
+  // ---- Model preference: positive cases ----

-  test("scoped 偏好 — 限定系列时 intent 含 modelPreference.mode=scoped", async () => {
+  test("scoped preference — intent contains modelPreference.mode=scoped when family is specified", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
      "--dry-run",
      "--message",
-      "deepseek系列中哪个模型最适合用来进行快速推理",
+      "Which model in the deepseek family is best for fast reasoning?",
      "--non-interactive",
      "--output",
      "json",
@@ -103,13 +107,13 @@ describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）",
    ).toBe(true);
  }, 60_000);

-  test("comparison 偏好 — 对比模型时 intent 含 modelPreference.mode=comparison", async () => {
+  test("comparison preference — intent contains modelPreference.mode=comparison when comparing models", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
      "--dry-run",
      "--message",
-      "qwen-max和deepseek-v3哪个更适合做代码生成",
+      "Which is better for code generation, qwen-max or deepseek-v3?",
      "--non-interactive",
      "--output",
      "json",
@@ -122,13 +126,13 @@ describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）",
    expect(data.intent?.modelPreference?.targets?.length).toBeGreaterThanOrEqual(2);
  }, 60_000);

-  test("excludes 偏好 — 排除模型时 intent 识别出 modelPreference", async () => {
+  test("excludes preference — intent detects modelPreference when excluding models", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
      "--dry-run",
      "--message",
-      "不要qwen，推荐一个适合文本生成的模型",
+      "Not qwen, recommend a model suitable for text generation",
      "--non-interactive",
      "--output",
      "json",
@@ -147,15 +151,15 @@ describe.skipIf(!isDashScopeE2EReady())("e2e: advisor recommend（DashScope）",
    expect(hasExcludes).toBe(true);
  }, 60_000);

-  // ---- 模型偏好：反例 ----
+  // ---- Model preference: negative cases ----

-  test("无偏好 — 普通需求查询时 intent 不含 modelPreference 或 mode=unconstrained", async () => {
+  test("no preference — intent has no modelPreference or mode=unconstrained for generic queries", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "advisor",
      "recommend",
      "--dry-run",
      "--message",
-      "我要做一个能理解图片的客服机器人",
+      "I want to build a customer service bot that understands images",
      "--non-interactive",
      "--output",
      "json",
@@ -17,6 +17,7 @@ describe("e2e: auth", () => {
    const { stderr, exitCode } = await runCli(["auth", "login", "--help"]);
    expect(exitCode, stderr).toBe(0);
    expect(stderr).toMatch(/login|api-key/i);
+    expect(stderr).toMatch(/--console-site/);
  });

  test("auth logout --help 正常退出", async () => {
@@ -159,20 +160,25 @@ describe("e2e: auth", () => {
    expect(data.dashscope_commands?.method).toBeDefined();
  });

-  test.skipIf(!isDashScopeE2EReady())("auth status --output json --quiet --region cn", async () => {
-    const { stdout, stderr, exitCode } = await runCli([
-      "auth",
-      "status",
-      "--non-interactive",
-      "--output",
-      "json",
-      "--quiet",
-      "--region",
-      "cn",
-    ]);
-    expect(exitCode, stderr).toBe(0);
-    const data = parseStdoutJson<{ authenticated?: boolean; dashscope_commands?: unknown }>(stdout);
-    expect(data.authenticated).toBe(true);
-    expect(data.dashscope_commands).toBeDefined();
-  });
+  test.skipIf(!isDashScopeE2EReady())(
+    "auth status --output json --quiet --base-url 国内",
+    async () => {
+      const { stdout, stderr, exitCode } = await runCli([
+        "auth",
+        "status",
+        "--non-interactive",
+        "--output",
+        "json",
+        "--quiet",
+        "--base-url",
+        "https://dashscope.aliyuncs.com",
+      ]);
+      expect(exitCode, stderr).toBe(0);
+      const data = parseStdoutJson<{ authenticated?: boolean; dashscope_commands?: unknown }>(
+        stdout,
+      );
+      expect(data.authenticated).toBe(true);
+      expect(data.dashscope_commands).toBeDefined();
+    },
+  );
 });
@@ -41,12 +41,10 @@ describe("e2e: config", () => {
    ]);
    expect(exitCode, stderr).toBe(0);
    const data = parseStdoutJson<{
-      region?: string;
      config_file?: string;
      base_url?: string;
      timeout?: number;
    }>(stdout);
-    expect(data.region).toBeDefined();
    expect(data.config_file).toBeDefined();
    expect(data.base_url).toBeDefined();
    expect(data.timeout).toBeDefined();
@@ -62,7 +60,7 @@ describe("e2e: config", () => {
      "--no-color",
    ]);
    expect(exitCode, stderr).toBe(0);
-    expect(stdout).toMatch(/region|config_file|timeout|base_url/i);
+    expect(stdout).toMatch(/config_file|timeout|base_url/i);
  });

  test("config set 缺少 --key / --value 时退出为用法错误 (2)", async () => {
@@ -85,20 +83,6 @@ describe("e2e: config", () => {
    expect(stderr).toMatch(/Invalid config key|not-a-real-key/i);
  });

-  test("config set 非法 region", async () => {
-    const { stderr, exitCode } = await runCli([
-      "config",
-      "set",
-      "--non-interactive",
-      "--key",
-      "region",
-      "--value",
-      "invalid-region",
-    ]);
-    expect(exitCode).toBe(2);
-    expect(stderr).toMatch(/Invalid region|cn, us, intl/i);
-  });
-
  test("config set 非法 output", async () => {
    const { stderr, exitCode } = await runCli([
      "config",
@@ -0,0 +1,148 @@
+import { describe, expect, test } from "vite-plus/test";
+import { parseStdoutJson, runCli } from "./helpers.ts";
+
+type ConsoleDryRunMeta = {
+  consoleRegion?: string;
+  consoleSite?: string;
+  consoleSwitchAgent?: number;
+};
+
+/**
+ * E2E for global console flags (`--console-region`, `--console-site`,
+ * `--console-switch-agent`) and DashScope `--base-url`.
+ */
+
+describe("e2e: console global flags", () => {
+  test("根帮助展示 --base-url 与 console 全局标志", async () => {
+    const { stderr, exitCode } = await runCli(["--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--base-url/);
+    expect(stderr).toMatch(/--console-region/);
+    expect(stderr).toMatch(/--console-site/);
+    expect(stderr).toMatch(/--console-switch-agent/);
+    expect(stderr).not.toMatch(/^\s*--region\s/m);
+  });
+
+  test("quota check --help 不重复命令级 region，并提示全局 flags", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "check", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/Global flags.*always available/i);
+    expect(stderr).toMatch(/--model <model>/);
+    expect(stderr).toMatch(/--period <minutes>/);
+    expect(stderr).not.toMatch(/API region \(default: cn-beijing\)/);
+  });
+
+  test("console call --help 不暴露命令级 region/site，示例使用 --console-region", async () => {
+    const { stderr, exitCode } = await runCli(["console", "call", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--api <api>/);
+    expect(stderr).toMatch(/--data <json>/);
+    expect(stderr).not.toMatch(/^\s*--region\s/m);
+    expect(stderr).not.toMatch(/^\s*--site\s/m);
+    expect(stderr).toMatch(/--console-region cn-beijing/);
+  });
+
+  test("auth login --help 描述 --console 与 --console-site 配合", async () => {
+    const { stderr, exitCode } = await runCli(["auth", "login", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--console-site/);
+    expect(stderr).toMatch(/--console.*console-site|console-site.*domestic|international/i);
+  });
+
+  test("console call --dry-run 默认 consoleRegion 为 cn-beijing", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "console",
+      "call",
+      "--api",
+      "some.api.name",
+      "--data",
+      "{}",
+      "--dry-run",
+      "--non-interactive",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<ConsoleDryRunMeta>(stdout);
+    expect(data.consoleRegion).toBe("cn-beijing");
+    expect(data.consoleSite).toBe("domestic");
+  });
+
+  test("console call --dry-run --console-region / --console-site / --console-switch-agent 透传", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "console",
+      "call",
+      "--api",
+      "some.api.name",
+      "--data",
+      "{}",
+      "--dry-run",
+      "--non-interactive",
+      "--output",
+      "json",
+      "--console-region",
+      "ap-southeast-1",
+      "--console-site",
+      "international",
+      "--console-switch-agent",
+      "12345",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<ConsoleDryRunMeta>(stdout);
+    expect(data.consoleRegion).toBe("ap-southeast-1");
+    expect(data.consoleSite).toBe("international");
+    expect(data.consoleSwitchAgent).toBe(12345);
+  });
+
+  test("console call 拒绝未知全局 flag --region", async () => {
+    const { stderr, exitCode } = await runCli([
+      "console",
+      "call",
+      "--api",
+      "some.api.name",
+      "--data",
+      "{}",
+      "--dry-run",
+      "--non-interactive",
+      "--region",
+      "cn",
+    ]);
+    expect(exitCode).not.toBe(0);
+    expect(stderr).toMatch(/Unknown flag.*--region/);
+  });
+
+  test("mcp list --dry-run --console-region 透传", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "mcp",
+      "list",
+      "--dry-run",
+      "--non-interactive",
+      "--output",
+      "json",
+      "--console-region",
+      "cn-hangzhou",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<ConsoleDryRunMeta>(stdout);
+    expect(data.consoleRegion).toBe("cn-hangzhou");
+  });
+
+  test("quota check --dry-run --console-region 透传", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--dry-run",
+      "--non-interactive",
+      "--output",
+      "json",
+      "--console-region",
+      "cn-hangzhou",
+      "--console-site",
+      "international",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<ConsoleDryRunMeta>(stdout);
+    expect(data.consoleRegion).toBe("cn-hangzhou");
+    expect(data.consoleSite).toBe("international");
+  });
+});
@@ -117,8 +117,17 @@ export function e2eLabelFromMetaUrl(metaUrl: string): string {
  return basename(fileURLToPath(metaUrl), ".ts").replace(/\.e2e\.test$/, "");
 }

-/** 知识库用例：须显式索引 ID + AK/SK（workspace 可读 config / env，故不在此强制校验） */
+/** 知识库用例：须显式索引 ID + API-KEY 或 AK/SK */
 export function isKnowledgeE2EReady(): boolean {
+  if (!isBailianE2EEnabled()) return false;
+  if (!process.env.BAILIAN_E2E_INDEX_ID) return false;
+  const hasApiKey = isDashScopeE2EReady();
+  const hasAkSk =
+    !!process.env.ALIBABA_CLOUD_ACCESS_KEY_ID && !!process.env.ALIBABA_CLOUD_ACCESS_KEY_SECRET;
+  return hasApiKey || hasAkSk;
+}
+
+export function isKnowledgeAkSkReady(): boolean {
  return (
    isBailianE2EEnabled() &&
    !!process.env.ALIBABA_CLOUD_ACCESS_KEY_ID &&
@@ -1,55 +1,206 @@
-import { join } from "path";
+import { tmpdir } from "os";
 import { describe, expect, test } from "vite-plus/test";
-import {
-  isBailianE2EEnabled,
-  isKnowledgeE2EReady,
-  monorepoRoot,
-  parseStdoutJson,
-  runCli,
-} from "./helpers.ts";
+import { parseStdoutJson, runCli } from "./helpers.ts";

-// 已开启 E2E 但 AK/SK、索引等未齐时提醒配置根目录 .env（否则本文件整组 describe 会被 skip）
-if (isBailianE2EEnabled() && !isKnowledgeE2EReady()) {
-  const envFile = join(monorepoRoot(), ".env");
-  console.warn(
-    [
-      "[e2e:knowledge] 知识库检索需要 RAM 的 AK/SK、索引 ID，以及工作空间 ID；当前未就绪，本组用例将被跳过。",
-      `请在 monorepo 根目录的 .env 中配置（${envFile}）：`,
-      "  ALIBABA_CLOUD_ACCESS_KEY_ID",
-      "  ALIBABA_CLOUD_ACCESS_KEY_SECRET",
-      "  BAILIAN_E2E_INDEX_ID",
-      "  BAILIAN_WORKSPACE_ID（也可执行: bl config set workspace_id <工作空间 id>）",
-    ].join("\n"),
-  );
+// ---- Types ----
+
+interface DryRunBody {
+  endpoint?: string;
+  request?: {
+    index_id?: string;
+    query?: string;
+    search_filters?: unknown[];
+    rerank_top_n?: number;
+    enable_reranking?: boolean;
+    dense_similarity_top_k?: number;
+    sparse_similarity_top_k?: number;
+    rerank?: Array<{ model_name?: string; rerank_mode?: string; rerank_instruct?: string }>;
+  };
 }

-interface KnowledgeRetrieveBody {
-  Success?: boolean;
-  Code?: string;
-  Data?: { Nodes?: unknown[] };
-}
+// ---- Help & missing args (no credentials needed) ----

-/** 知识库检索（需 AK/SK + workspace + 索引；未就绪则整组跳过） */
-describe.skipIf(!isKnowledgeE2EReady())("e2e: knowledge retrieve", () => {
-  test("知识库检索", async () => {
-    const indexId = process.env.BAILIAN_E2E_INDEX_ID!;
-    const { stdout, stderr, exitCode } = await runCli([
+describe("e2e: knowledge retrieve", () => {
+  test("knowledge 分组展示子命令帮助且成功退出", async () => {
+    const { stdout, stderr, exitCode } = await runCli(["knowledge"]);
+    expect(exitCode, stderr).toBe(0);
+    const out = `${stdout}\n${stderr}`;
+    expect(out).toMatch(/knowledge|retrieve/i);
+  });
+
+  test("knowledge retrieve --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["knowledge", "retrieve", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--index-id/i);
+    expect(stderr).toMatch(/--query/i);
+    expect(stderr).toMatch(/--rerank-top-n/i);
+    expect(stderr).toMatch(/deprecated/i);
+    expect(stderr).toMatch(/--workspace-id/i);
+  });
+
+  test("缺少 --index-id 时打印帮助并退出 (0)", async () => {
+    const { stderr, exitCode } = await runCli([
+      "knowledge",
+      "retrieve",
+      "--query",
+      "test",
+      "--non-interactive",
+    ]);
+    expect(exitCode).toBe(0);
+    expect(stderr).toMatch(/--index-id|Usage:/i);
+  });
+
+  test("缺少 --query 时打印帮助并退出 (0)", async () => {
+    const { stderr, exitCode } = await runCli([
      "knowledge",
      "retrieve",
      "--index-id",
-      indexId,
-      "--query",
-      "端到端检索测试",
-      "--top-k",
-      "3",
+      "idx_test",
      "--non-interactive",
-      "--output",
-      "json",
    ]);
-    expect(exitCode, stderr).toBe(0);
-    const data = parseStdoutJson<KnowledgeRetrieveBody>(stdout);
-    const ok = data.Success === true || data.Code === "Success";
-    expect(ok).toBe(true);
-    expect(Array.isArray(data.Data?.Nodes)).toBe(true);
-  }, 120_000);
+    expect(exitCode).toBe(0);
+    expect(stderr).toMatch(/--query|Usage:/i);
+  });
+});
+
+// ---- Error scenarios (no real credentials needed) ----
+
+describe("e2e: knowledge retrieve errors", () => {
+  test("无任何凭证时提示 No credentials found 并非零退出", async () => {
+    const { stderr, exitCode } = await runCli(
+      [
+        "knowledge",
+        "retrieve",
+        "--index-id",
+        "idx_test",
+        "--query",
+        "test",
+        "--non-interactive",
+        "--output",
+        "json",
+      ],
+      {
+        DASHSCOPE_API_KEY: undefined,
+        DASHSCOPE_ACCESS_TOKEN: undefined,
+        ALIBABA_CLOUD_ACCESS_KEY_ID: undefined,
+        ALIBABA_CLOUD_ACCESS_KEY_SECRET: undefined,
+        BAILIAN_CONFIG_DIR: tmpdir(),
+      },
+    );
+    expect(exitCode).not.toBe(0);
+    expect(stderr).toMatch(/no credentials found/i);
+  });
+});
+
+// ---- Dry-run (no real credentials needed) ----
+
+describe("e2e: knowledge retrieve dry-run", () => {
+  test("--dry-run 输出 endpoint 和 snake_case body", async () => {
+    const { stdout, stderr, exitCode } = await runCli(
+      [
+        "knowledge",
+        "retrieve",
+        "--dry-run",
+        "--index-id",
+        "idx_test",
+        "--query",
+        "hello",
+        "--non-interactive",
+        "--output",
+        "json",
+      ],
+      { DASHSCOPE_API_KEY: "sk-fake-for-dryrun" },
+    );
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<DryRunBody>(stdout);
+    expect(data.endpoint).toMatch(/api\/v1\/indices\/rag\/index\/retrieve/);
+    expect(data.request?.index_id).toBe("idx_test");
+    expect(data.request?.query).toBe("hello");
+  });
+
+  test("--dry-run + --top-k 转发到 rerank_top_n 并输出废弃警告", async () => {
+    const { stdout, stderr, exitCode } = await runCli(
+      [
+        "knowledge",
+        "retrieve",
+        "--dry-run",
+        "--index-id",
+        "idx_test",
+        "--query",
+        "hello",
+        "--top-k",
+        "5",
+        "--non-interactive",
+        "--output",
+        "json",
+      ],
+      { DASHSCOPE_API_KEY: "sk-fake-for-dryrun" },
+    );
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--top-k.*deprecated/i);
+    const data = parseStdoutJson<DryRunBody>(stdout);
+    expect(data.request?.rerank_top_n).toBe(5);
+  });
+
+  test("--dry-run + --rerank-top-n 优先于 --top-k", async () => {
+    const { stdout, stderr, exitCode } = await runCli(
+      [
+        "knowledge",
+        "retrieve",
+        "--dry-run",
+        "--index-id",
+        "idx_test",
+        "--query",
+        "hello",
+        "--top-k",
+        "5",
+        "--rerank-top-n",
+        "10",
+        "--non-interactive",
+        "--output",
+        "json",
+      ],
+      { DASHSCOPE_API_KEY: "sk-fake-for-dryrun" },
+    );
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<DryRunBody>(stdout);
+    expect(data.request?.rerank_top_n).toBe(10);
+  });
+
+  test("--dry-run + rerank 参数完整输出", async () => {
+    const { stdout, stderr, exitCode } = await runCli(
+      [
+        "knowledge",
+        "retrieve",
+        "--dry-run",
+        "--index-id",
+        "idx_test",
+        "--query",
+        "hello",
+        "--rerank",
+        "--rerank-model",
+        "qwen3-rerank-hybrid",
+        "--rerank-mode",
+        "custom",
+        "--rerank-instruct",
+        "按相关性排序",
+        "--dense-similarity-top-k",
+        "100",
+        "--sparse-similarity-top-k",
+        "50",
+        "--non-interactive",
+        "--output",
+        "json",
+      ],
+      { DASHSCOPE_API_KEY: "sk-fake-for-dryrun" },
+    );
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<DryRunBody>(stdout);
+    expect(data.request?.enable_reranking).toBe(true);
+    expect(data.request?.dense_similarity_top_k).toBe(100);
+    expect(data.request?.sparse_similarity_top_k).toBe(50);
+    expect(data.request?.rerank?.[0]?.model_name).toBe("qwen3-rerank-hybrid");
+    expect(data.request?.rerank?.[0]?.rerank_mode).toBe("custom");
+    expect(data.request?.rerank?.[0]?.rerank_instruct).toBe("按相关性排序");
+  });
 });
@@ -66,7 +66,7 @@ describe("e2e: mcp", () => {
    expect(exitCode, stderr).toBe(0);
    const data = parseStdoutJson<{
      api?: string;
-      region?: string;
+      consoleRegion?: string;
      data?: {
        reqDTO?: {
          type?: string;
@@ -79,7 +79,6 @@ describe("e2e: mcp", () => {
      };
    }>(stdout);
    expect(data.api).toBe("zeldaEasy.broadscope-bailian.mcp-server.PageList");
-    expect(data.region).toBe("cn-beijing");
    expect(data.data?.reqDTO?.activated).toBe(1);
    expect(data.data?.reqDTO?.displayTools).toBe(false);
    expect(data.data?.reqDTO?.type).toBe("OFFICIAL");
@@ -88,7 +87,7 @@ describe("e2e: mcp", () => {
    expect(data.data?.reqDTO?.pageSize).toBe(5);
  });

-  test("mcp list --dry-run 自定义 --region 透传", async () => {
+  test("mcp list --dry-run 自定义 --console-region 透传", async () => {
    const { stdout, stderr, exitCode } = await runCli([
      "mcp",
      "list",
@@ -96,12 +95,12 @@ describe("e2e: mcp", () => {
      "--non-interactive",
      "--output",
      "json",
-      "--region",
+      "--console-region",
      "cn-hangzhou",
    ]);
    expect(exitCode, stderr).toBe(0);
-    const data = parseStdoutJson<{ region?: string }>(stdout);
-    expect(data.region).toBe("cn-hangzhou");
+    const data = parseStdoutJson<{ consoleRegion?: string }>(stdout);
+    expect(data.consoleRegion).toBe("cn-hangzhou");
  });

  test("mcp tools <server-code> --dry-run 输出 /api/v1/mcps/<code>/mcp 形态 URL", async () => {
@@ -0,0 +1,139 @@
+import { describe, expect, test } from "vite-plus/test";
+import { join } from "node:path";
+import {
+  e2eLabelFromMetaUrl,
+  isBailianE2EMediaEnabled,
+  isDashScopeE2EReady,
+  makeE2eOutputDir,
+  parseStdoutJson,
+  runCli,
+} from "./helpers.ts";
+
+describe("e2e: omni", () => {
+  test("omni --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["omni", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/omni|--message|--audio|text-only/i);
+  });
+});
+
+describe.skipIf(!isBailianE2EMediaEnabled() || !isDashScopeE2EReady())(
+  "e2e: omni（DashScope 媒体）",
+  () => {
+    test("omni --list-voices 输出音色列表并退出", async () => {
+      const { stdout, stderr, exitCode } = await runCli(["omni", "--list-voices"]);
+      expect(exitCode, stderr).toBe(0);
+      expect(stdout).toMatch(/Omni output voices:/);
+      expect(stdout).toMatch(/Tina/);
+      expect(stdout).toMatch(/Dylan/);
+      expect(stdout).toMatch(/Total: 13 voices/);
+    });
+    test("omni 缺少 --message 时打印子命令帮助并退出 (0)", async () => {
+      const { stderr, exitCode } = await runCli([
+        "omni",
+        "--model",
+        "qwen3.5-omni-flash",
+        "--non-interactive",
+      ]);
+      expect(exitCode).toBe(0);
+      expect(stderr).toMatch(/--message|Usage:/i);
+    });
+
+    test("omni --audio 无法识别扩展名时退出为用法错误 (2)", async () => {
+      const { stderr, exitCode } = await runCli([
+        "omni",
+        "--model",
+        "qwen3.5-omni-flash",
+        "--audio",
+        "https://example.com/sample.flac",
+        "--text-only",
+        "--message",
+        "这段音频在说什么？",
+        "--non-interactive",
+      ]);
+      expect(exitCode).toBe(2);
+      expect(stderr).toMatch(/Unsupported audio extension|Cannot infer audio format/i);
+    });
+
+    test("omni --dry-run --audio 构造 input_audio 而非 audio_url", async () => {
+      const { stdout, stderr, exitCode } = await runCli([
+        "omni",
+        "--dry-run",
+        "--model",
+        "qwen3.5-omni-flash",
+        "--audio",
+        "https://example.com/sample.wav",
+        "--text-only",
+        "--message",
+        "这段音频在说什么？",
+        "--non-interactive",
+        "--output",
+        "json",
+      ]);
+      expect(exitCode, stderr).toBe(0);
+      const data = parseStdoutJson<{
+        request?: {
+          messages?: Array<{
+            content?: Array<{
+              type?: string;
+              audio_url?: unknown;
+              input_audio?: { data?: string; format?: string };
+            }>;
+          }>;
+        };
+      }>(stdout);
+      const parts = data.request?.messages?.flatMap((m) =>
+        Array.isArray(m.content) ? m.content : [],
+      );
+      const audioPart = parts?.find((p) => p.type === "input_audio" || p.type === "audio_url");
+      expect(audioPart?.type).toBe("input_audio");
+      expect(audioPart?.audio_url).toBeUndefined();
+      expect(audioPart?.input_audio?.data).toBe("https://example.com/sample.wav");
+      expect(audioPart?.input_audio?.format).toBe("wav");
+    });
+
+    test("【qwen3.5-omni-flash】本地音频理解", async () => {
+      const outDir = makeE2eOutputDir(e2eLabelFromMetaUrl(import.meta.url));
+      const clipText = "端到端Omni音频测试";
+      const clipWav = join(outDir, "e2e-omni-input.wav");
+
+      const syn = await runCli([
+        "speech",
+        "synthesize",
+        "--model",
+        "cosyvoice-v3-flash",
+        "--voice",
+        "longxiaochun_v3",
+        "--text",
+        clipText,
+        "--format",
+        "wav",
+        "--out",
+        clipWav,
+        "--non-interactive",
+        "--output",
+        "json",
+      ]);
+      expect(syn.exitCode, syn.stderr).toBe(0);
+
+      const omni = await runCli([
+        "omni",
+        "--model",
+        "qwen3.5-omni-flash",
+        "--audio",
+        clipWav,
+        "--text-only",
+        "--system",
+        "请逐字转写用户提供的音频内容，不要添加解释。",
+        "--message",
+        "请转写这段音频。",
+        "--non-interactive",
+        "--output",
+        "json",
+      ]);
+      expect(omni.exitCode, omni.stderr).toBe(0);
+      const body = parseStdoutJson<{ content?: string }>(omni.stdout);
+      expect(body.content?.replace(/\s/g, "")).toMatch(/端到端Omni音频测试/);
+    }, 180_000);
+  },
+);
@@ -0,0 +1,124 @@
+import { execFile } from "child_process";
+import { createServer, type Server } from "http";
+import { mkdtempSync, rmSync, writeFileSync } from "fs";
+import type { AddressInfo } from "net";
+import { tmpdir } from "os";
+import { join } from "path";
+import { promisify } from "util";
+import { afterAll, beforeAll, describe, expect, test } from "vite-plus/test";
+import { cliPackageRoot } from "./helpers.ts";
+
+const execFileAsync = promisify(execFile);
+
+/**
+ * 代理支持 E2E（issue #35）：只验证 `setupProxyFromEnv()` 是否把代理 dispatcher
+ * 正确装到全局 fetch 上——设了 HTTPS_PROXY 后裸 `fetch()` 走代理，未设置时直连，
+ * NO_PROXY 命中时跳过，非法代理值给出明确报错。
+ *
+ * 不经过任何 CLI 命令（不解析凭证、不打 gateway），因此 CI 上无需 api key /
+ * access token，与既有 e2e 设计一致。全程离线：目标域名用 `.invalid`（保留顶级域，
+ * 必然无法解析），代理收到 CONNECT 后规范返回 502，不产生真实外网请求。
+ */
+
+const FAKE_HOST = "bl-proxy-e2e.invalid";
+const FAKE_URL = `https://${FAKE_HOST}/probe`;
+
+/**
+ * 最小探针脚本：调用真实的 `setupProxyFromEnv()`，再对目标发一个普通 fetch。
+ * 代理行为由进程环境变量决定，正是被测对象；fetch 成败不重要，我们只看代理是否收到 CONNECT。
+ */
+const PROBE_SCRIPT = `
+import { setupProxyFromEnv } from ${JSON.stringify(join(cliPackageRoot, "src", "proxy.ts"))};
+setupProxyFromEnv();
+try {
+  await fetch(${JSON.stringify(FAKE_URL)}, { signal: AbortSignal.timeout(5000) });
+} catch {
+  // 目标不可达/隧道被拒都正常——本测试只关心代理是否收到 CONNECT
+}
+`;
+
+let proxy: Server;
+let proxyUrl: string;
+let scriptDir: string;
+let scriptPath: string;
+const connectTargets: string[] = [];
+
+beforeAll(async () => {
+  proxy = createServer();
+  // 记录收到的 CONNECT 目标（host:port），并以 502 拒绝隧道
+  proxy.on("connect", (req, clientSocket) => {
+    connectTargets.push(req.url ?? "");
+    clientSocket.end("HTTP/1.1 502 Bad Gateway\r\n\r\n");
+  });
+  await new Promise<void>((resolve) => proxy.listen(0, "127.0.0.1", resolve));
+  proxyUrl = `http://127.0.0.1:${(proxy.address() as AddressInfo).port}`;
+
+  scriptDir = mkdtempSync(join(tmpdir(), "bl-proxy-e2e-"));
+  scriptPath = join(scriptDir, "probe.ts");
+  writeFileSync(scriptPath, PROBE_SCRIPT);
+});
+
+afterAll(async () => {
+  await new Promise<void>((resolve) => proxy.close(() => resolve()));
+  rmSync(scriptDir, { recursive: true, force: true });
+});
+
+/** 清空所有代理相关环境变量，确保每个用例只受自身设置影响 */
+const PROXY_ENV_CLEARED = {
+  HTTPS_PROXY: "",
+  https_proxy: "",
+  HTTP_PROXY: "",
+  http_proxy: "",
+  NO_PROXY: "",
+  no_proxy: "",
+};
+
+/** 以给定代理环境变量运行探针脚本，返回 { exitCode, stderr } */
+async function runProbe(
+  envOverrides: NodeJS.ProcessEnv,
+): Promise<{ exitCode: number; stderr: string }> {
+  try {
+    await execFileAsync("node", [scriptPath], {
+      cwd: cliPackageRoot,
+      encoding: "utf8",
+      env: { ...process.env, NODE_NO_WARNINGS: "1", ...PROXY_ENV_CLEARED, ...envOverrides },
+    });
+    return { exitCode: 0, stderr: "" };
+  } catch (err: unknown) {
+    const e = err as { stderr?: string; code?: number };
+    return { exitCode: typeof e.code === "number" ? e.code : 1, stderr: e.stderr ?? "" };
+  }
+}
+
+describe("e2e: proxy", () => {
+  test("设置 HTTPS_PROXY 后 fetch 经过代理（CONNECT 到目标主机）", async () => {
+    connectTargets.length = 0;
+    await runProbe({ HTTPS_PROXY: proxyUrl });
+    expect(connectTargets).toContain(`${FAKE_HOST}:443`);
+  });
+
+  test("空字符串小写变量不屏蔽大写 HTTPS_PROXY（undici ?? 取值回归）", async () => {
+    connectTargets.length = 0;
+    await runProbe({ https_proxy: "", HTTPS_PROXY: proxyUrl });
+    expect(connectTargets).toContain(`${FAKE_HOST}:443`);
+  });
+
+  test("NO_PROXY 命中目标主机时不走代理", async () => {
+    connectTargets.length = 0;
+    await runProbe({ HTTPS_PROXY: proxyUrl, NO_PROXY: FAKE_HOST });
+    expect(connectTargets.filter((t) => t.startsWith(FAKE_HOST))).toEqual([]);
+  });
+
+  test("未设置代理变量时保持直连（代理收不到任何流量）", async () => {
+    connectTargets.length = 0;
+    await runProbe({});
+    expect(connectTargets).toEqual([]);
+  });
+
+  test("代理 URL 非法时给出明确报错而非堆栈", async () => {
+    const { exitCode, stderr } = await runProbe({ HTTPS_PROXY: "::::not-a-url" });
+    expect(exitCode).not.toBe(0);
+    expect(stderr).toMatch(/Invalid proxy configuration/);
+    expect(stderr).toMatch(/HTTPS_PROXY/);
+  });
+});
@@ -0,0 +1,367 @@
+import { describe, expect, test } from "vite-plus/test";
+import { isBailianE2EEnabled, parseStdoutJson, runCli } from "./helpers.ts";
+import { readConfigFile } from "bailian-cli-core";
+
+function isConsoleE2EReady(): boolean {
+  if (!isBailianE2EEnabled()) return false;
+  if (process.env.DASHSCOPE_ACCESS_TOKEN?.trim()) return true;
+  try {
+    const config = readConfigFile();
+    return typeof config.access_token === "string" && config.access_token.length > 0;
+  } catch {
+    return false;
+  }
+}
+
+describe("e2e: quota", () => {
+  test("quota list --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "list", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("--model");
+    expect(stderr).toContain("--all");
+  });
+
+  test("quota list --help 包含所有示例", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "list", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("bl quota list");
+    expect(stderr).toContain("bl quota list --model qwen3.6-plus");
+    expect(stderr).toContain("bl quota list --all");
+  });
+
+  test("quota request --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "request", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("--model");
+    expect(stderr).toContain("--tpm");
+    expect(stderr).toContain("--yes");
+  });
+
+  test("quota history --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "history", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("--page");
+    expect(stderr).toContain("--model");
+  });
+
+  test("quota check --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "check", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("--model");
+    expect(stderr).toContain("--period");
+    expect(stderr).toContain("bl quota check");
+  });
+
+  test("quota check --period 0 报错最小值", async () => {
+    const { stderr, exitCode } = await runCli(["quota", "check", "--period", "0.5"]);
+    expect(exitCode).toBe(1);
+    expect(stderr).toContain("at least 1 minute");
+  });
+});
+
+describe.skipIf(!isConsoleE2EReady())("e2e: quota（Console）", () => {
+  test("quota list --dry-run 输出请求参数", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "list",
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      api?: string;
+      data?: {
+        input?: { queryQpmInfo?: boolean; supports?: { selfServiceLimitIncrease?: boolean } };
+      };
+    }>(stdout);
+    expect(data.api).toContain("listFoundationModels");
+    expect(data.data?.input?.queryQpmInfo).toBe(true);
+    expect(data.data?.input?.supports?.selfServiceLimitIncrease).toBe(true);
+  });
+
+  test("quota list --dry-run --all 不传 supports 过滤", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "list",
+      "--all",
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { input?: { supports?: unknown } };
+    }>(stdout);
+    expect(data.data?.input?.supports).toBeUndefined();
+  });
+
+  test("quota list 文本输出包含英文表头", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "list",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("Model");
+    expect(stdout).toContain("Req/min");
+    expect(stdout).toContain("Token/min");
+    expect(stdout).toContain("Max TPM");
+  });
+
+  test("quota list --model 指定模型返回结果", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "list",
+      "--model",
+      "qwen3.6-plus",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("qwen3.6-plus");
+    expect(stdout).toMatch(/Total: 1 models/);
+  });
+
+  test("quota list --model 不存在的模型报错", async () => {
+    const { stderr, exitCode } = await runCli([
+      "quota",
+      "list",
+      "--model",
+      "nonexistent-model-xyz-99999",
+      "--output",
+      "text",
+    ]);
+    expect(exitCode).toBe(1);
+    expect(stderr).toContain("no matching models found");
+  });
+
+  test("quota list JSON 输出包含 model/rpm/tpm/maxTPM", async () => {
+    const { stdout, stderr, exitCode } = await runCli(["quota", "list", "--output", "json"]);
+    expect(exitCode, stderr).toBe(0);
+    const data =
+      parseStdoutJson<
+        Array<{ model?: string; rpm?: number | null; tpm?: number | null; maxTPM?: number | null }>
+      >(stdout);
+    expect(Array.isArray(data)).toBe(true);
+    expect(data.length).toBeGreaterThan(0);
+    expect(data[0].model).toBeTypeOf("string");
+    expect(data[0].rpm).toBeTypeOf("number");
+    expect(data[0].tpm).toBeTypeOf("number");
+    expect(data[0].maxTPM).toBeTypeOf("number");
+  });
+
+  test("quota request --dry-run 输出请求参数", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "request",
+      "--model",
+      "qwen3.6-plus",
+      "--tpm",
+      "6000000",
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      api?: string;
+      data?: { input?: { model?: string; limit?: { usage_limit?: number } } };
+    }>(stdout);
+    expect(data.api).toContain("updateFoundationModelLimits");
+    expect(data.data?.input?.model).toBe("qwen3.6-plus");
+    expect(data.data?.input?.limit?.usage_limit).toBeTypeOf("number");
+  });
+
+  test("quota request TPM 超范围报错", async () => {
+    const { stderr, exitCode } = await runCli([
+      "quota",
+      "request",
+      "--model",
+      "qwen3.6-plus",
+      "--tpm",
+      "999",
+    ]);
+    expect(exitCode).toBe(1);
+    expect(stderr).toContain("out of range");
+    expect(stderr).toContain("Current");
+    expect(stderr).toContain("Range");
+  });
+
+  test("quota request 不支持提额的模型报错", async () => {
+    const { stderr, exitCode } = await runCli([
+      "quota",
+      "request",
+      "--model",
+      "nonexistent-model-xyz-99999",
+      "--tpm",
+      "100000",
+    ]);
+    expect(exitCode).toBe(1);
+    expect(stderr).toContain("not found");
+  });
+
+  test("quota history --dry-run 输出请求参数", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "history",
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      api?: string;
+      data?: { input?: { pageNo?: number; pageSize?: number } };
+    }>(stdout);
+    expect(data.api).toContain("listModelLimitApplications");
+    expect(data.data?.input?.pageNo).toBe(1);
+    expect(data.data?.input?.pageSize).toBe(10);
+  });
+
+  test("quota check --dry-run 输出 API 信息", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{ apis?: string[]; consoleRegion?: string }>(stdout);
+    expect(data.apis).toContain(
+      "zeldaHttp.dashscopeModel./zelda/api/v1/modelCenter/listFoundationModels",
+    );
+    expect(data.apis).toContain("zeldaEasy.bailian-telemetry.monitor.getMonitorData");
+    expect(data.consoleRegion).toBe("cn-beijing");
+  });
+
+  test("quota check --dry-run --console-region 透传", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--dry-run",
+      "--non-interactive",
+      "--output",
+      "json",
+      "--console-region",
+      "cn-hangzhou",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{ consoleRegion?: string }>(stdout);
+    expect(data.consoleRegion).toBe("cn-hangzhou");
+  });
+
+  test("quota check 文本输出包含英文表头", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("Model");
+    expect(stdout).toContain("RPM Usage/Limit");
+    expect(stdout).toContain("TPM Usage/Limit");
+    expect(stdout).toContain("Status");
+  });
+
+  test("quota check --model 指定单模型", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--model",
+      "qwen3.6-plus",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("qwen3.6-plus");
+    expect(stdout).toMatch(/Total: 1 models/);
+  });
+
+  test("quota check --model 逗号分隔多模型", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--model",
+      "qwen3.6-plus,qwen-plus",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("qwen3.6-plus");
+    expect(stdout).toContain("qwen-plus");
+    expect(stdout).toMatch(/Total: 2 models/);
+  });
+
+  test("quota check JSON 输出包含用量和限额字段", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--model",
+      "qwen3.6-plus",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<
+      Array<{
+        model?: string;
+        rpmUsage?: number;
+        rpmLimit?: number;
+        tpmUsage?: number;
+        tpmLimit?: number;
+      }>
+    >(stdout);
+    expect(Array.isArray(data)).toBe(true);
+    expect(data.length).toBe(1);
+    expect(data[0].model).toBe("qwen3.6-plus");
+    expect(data[0].rpmUsage).toBeTypeOf("number");
+    expect(data[0].rpmLimit).toBeTypeOf("number");
+    expect(data[0].tpmUsage).toBeTypeOf("number");
+    expect(data[0].tpmLimit).toBeTypeOf("number");
+  });
+
+  test("quota check 状态列显示 Normal/Near limit/Rate Limited 之一", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "check",
+      "--model",
+      "qwen3.6-plus",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const hasStatus =
+      stdout.includes("Normal") || stdout.includes("Near limit") || stdout.includes("Rate Limited");
+    expect(hasStatus).toBe(true);
+  });
+
+  test("quota history --dry-run --page 2 --page-size 20", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "quota",
+      "history",
+      "--page",
+      "2",
+      "--page-size",
+      "20",
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { input?: { pageNo?: number; pageSize?: number } };
+    }>(stdout);
+    expect(data.data?.input?.pageNo).toBe(2);
+    expect(data.data?.input?.pageSize).toBe(20);
+  });
+});
@@ -0,0 +1,282 @@
+import { describe, expect, test } from "vite-plus/test";
+import { isBailianE2EEnabled, parseStdoutJson, runCli } from "./helpers.ts";
+import { readConfigFile } from "bailian-cli-core";
+
+function isConsoleE2EReady(): boolean {
+  if (!isBailianE2EEnabled()) return false;
+  if (process.env.DASHSCOPE_ACCESS_TOKEN?.trim()) return true;
+  try {
+    const config = readConfigFile();
+    return typeof config.access_token === "string" && config.access_token.length > 0;
+  } catch {
+    return false;
+  }
+}
+
+describe("e2e: usage free", () => {
+  test("usage 分组展示子命令帮助且退出码为 0", async () => {
+    const { stdout, stderr, exitCode } = await runCli(["usage"]);
+    expect(exitCode, stderr).toBe(0);
+    const out = `${stdout}\n${stderr}`;
+    expect(out).toMatch(/usage|free|freetier/i);
+  });
+
+  test("usage free --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["usage", "free", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--model|quota|free-tier/i);
+  });
+
+  test("usage free --help 包含所有示例", async () => {
+    const { stderr, exitCode } = await runCli(["usage", "free", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("bl usage free");
+    expect(stderr).toContain("bl usage free --model qwen3-max");
+    expect(stderr).toContain("bl usage free --model qwen3-max,qwen-turbo");
+  });
+});
+
+describe.skipIf(!isConsoleE2EReady())("e2e: usage free（Console）", () => {
+  test("usage free --dry-run --model 输出请求参数不发起调用", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--dry-run",
+      "--model",
+      "qwen3-max",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      api?: string;
+      data?: { queryFreeTierQuotaRequest?: { models?: string[] } };
+    }>(stdout);
+    expect(data.api).toContain("queryFreeTierQuota");
+    expect(data.data?.queryFreeTierQuotaRequest?.models).toEqual(["qwen3-max"]);
+  });
+
+  test("usage free --dry-run --model 逗号分隔多个模型", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--dry-run",
+      "--model",
+      "qwen3-max,qwen-turbo",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { queryFreeTierQuotaRequest?: { models?: string[] } };
+    }>(stdout);
+    expect(data.data?.queryFreeTierQuotaRequest?.models).toEqual(["qwen3-max", "qwen-turbo"]);
+  });
+
+  test("usage free --dry-run --model 重复模型名自动去重", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--dry-run",
+      "--model",
+      "qwen3-max,qwen3-max,qwen-turbo",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { queryFreeTierQuotaRequest?: { models?: string[] } };
+    }>(stdout);
+    expect(data.data?.queryFreeTierQuotaRequest?.models).toEqual(["qwen3-max", "qwen-turbo"]);
+  });
+
+  test("usage free --dry-run --model 逗号间有空格也能正确解析", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--dry-run",
+      "--model",
+      "qwen3-max, qwen-turbo",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { queryFreeTierQuotaRequest?: { models?: string[] } };
+    }>(stdout);
+    expect(data.data?.queryFreeTierQuotaRequest?.models).toEqual(["qwen3-max", "qwen-turbo"]);
+  });
+
+  test("usage free --dry-run 不指定 --model 传全量模型列表", async () => {
+    const { stderr, exitCode } = await runCli(["usage", "free", "--dry-run", "--output", "json"]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage free --model 单模型查询返回 JSON 结果", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<
+      Array<{
+        model?: string;
+        type?: string | null;
+        remaining?: number | null;
+        total?: number | null;
+        usagePercent?: number | null;
+        expires?: string | null;
+        autoStop?: boolean | string | null;
+      }>
+    >(stdout);
+    expect(Array.isArray(data)).toBe(true);
+    expect(data.length).toBeGreaterThan(0);
+    expect(data[0].model).toBe("qwen3-max");
+    expect(data[0].type).toBeTypeOf("string");
+  });
+
+  test("usage free --model 单模型文本输出包含表头", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("Model");
+    expect(stdout).toContain("Type");
+    expect(stdout).toContain("Remaining/Total");
+    expect(stdout).toContain("Usage");
+    expect(stdout).toContain("Expires");
+    expect(stdout).toContain("Auto-Stop");
+  });
+
+  test("usage free --model 文本输出包含模型名", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("qwen3-max");
+  });
+
+  test("usage free --model 逗号分隔多模型文本输出包含所有模型", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max,qwen-turbo",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("qwen3-max");
+    expect(stdout).toContain("qwen-turbo");
+  });
+
+  test("usage free --model 文本输出包含正确的 Type 列", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("Text");
+  });
+
+  test("usage free --model quotaStatus 为 UNKNOWN 时 Auto-Stop 显示 Unsupported", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "wan2.7-image",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("Unsupported");
+  });
+
+  test("usage free --model quotaStatus 为 UNKNOWN 时额度显示为 -", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "wan2.7-image",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const lines = stdout.split("\n").filter((line) => line.includes("wan2.7-image"));
+    expect(lines.length).toBe(1);
+    expect(lines[0]).toContain("Vision");
+    expect(lines[0]).toContain("Unsupported");
+  });
+
+  test("usage free --model 不存在的模型仍返回表格行", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "nonexistent-model-xyz-12345",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stdout).toContain("nonexistent-model-xyz-12345");
+  });
+
+  test("usage free --model Auto-Stop 显示 ON、OFF 或 Unsupported", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const hasAutoStop =
+      stdout.includes("ON") || stdout.includes("OFF") || stdout.includes("Unsupported");
+    expect(hasAutoStop).toBe(true);
+  });
+
+  test("usage free --model --console-region cn-beijing 指定区域查询", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "free",
+      "--model",
+      "qwen3-max",
+      "--console-region",
+      "cn-beijing",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<Array<{ model?: string }>>(stdout);
+    expect(Array.isArray(data)).toBe(true);
+    expect(data.length).toBeGreaterThan(0);
+    expect(data[0].model).toBe("qwen3-max");
+  });
+});
@@ -0,0 +1,273 @@
+import { describe, expect, test } from "vite-plus/test";
+import { isBailianE2EEnabled, parseStdoutJson, runCli } from "./helpers.ts";
+import { readConfigFile } from "bailian-cli-core";
+
+function isConsoleE2EReady(): boolean {
+  if (!isBailianE2EEnabled()) return false;
+  if (process.env.DASHSCOPE_ACCESS_TOKEN?.trim()) return true;
+  try {
+    const config = readConfigFile();
+    return typeof config.access_token === "string" && config.access_token.length > 0;
+  } catch {
+    return false;
+  }
+}
+
+function getStaticWorkspaceId(): string | undefined {
+  if (process.env.BAILIAN_WORKSPACE_ID?.trim()) return process.env.BAILIAN_WORKSPACE_ID.trim();
+  try {
+    const config = readConfigFile();
+    if (config.workspace_id) return config.workspace_id;
+  } catch {}
+  return undefined;
+}
+
+async function fetchDefaultWorkspaceId(): Promise<string> {
+  const staticId = getStaticWorkspaceId();
+  if (staticId) return staticId;
+
+  const { stdout } = await runCli(["workspace", "list", "--output", "json"]);
+  const result = JSON.parse(stdout);
+  const data = result?.data?.DataV2?.data?.data?.data ?? [];
+  const defaultWs = data.find((ws: { defaultAgent?: boolean }) => ws.defaultAgent);
+  if (defaultWs?.workspaceId) return defaultWs.workspaceId;
+  if (data.length > 0 && data[0].workspaceId) return data[0].workspaceId;
+  throw new Error("No workspace found for e2e tests");
+}
+
+describe("e2e: usage stats", () => {
+  test("usage stats --help 正常退出", async () => {
+    const { stderr, exitCode } = await runCli(["usage", "stats", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toMatch(/--model|--days|stats/i);
+  });
+
+  test("usage stats --help 包含所有示例", async () => {
+    const { stderr, exitCode } = await runCli(["usage", "stats", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("bl usage stats");
+    expect(stderr).toContain("bl usage stats --model qwen-turbo");
+    expect(stderr).toContain("bl usage stats --days 30");
+  });
+
+  test("usage stats --help 包含 --workspace-id 选项", async () => {
+    const { stderr, exitCode } = await runCli(["usage", "stats", "--help"]);
+    expect(exitCode, stderr).toBe(0);
+    expect(stderr).toContain("--workspace-id");
+  });
+});
+
+describe.skipIf(!isConsoleE2EReady())("e2e: usage stats（Console）", () => {
+  let wsId: string;
+
+  test("获取默认 workspace-id", async () => {
+    wsId = await fetchDefaultWorkspaceId();
+    expect(wsId).toBeTypeOf("string");
+    expect(wsId.length).toBeGreaterThan(0);
+  });
+
+  test("usage stats --dry-run 概览模式输出请求参数", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--dry-run",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      api?: string;
+      data?: {
+        reqDTO?: {
+          startTime?: number;
+          endTime?: number;
+          modelCallSource?: string;
+          filterWorkspaceId?: string;
+        };
+      };
+    }>(stdout);
+    expect(data.api).toContain("getModelUsageStatistic");
+    expect(data.data?.reqDTO?.modelCallSource).toBe("Online");
+    expect(data.data?.reqDTO?.startTime).toBeTypeOf("number");
+    expect(data.data?.reqDTO?.endTime).toBeTypeOf("number");
+    expect(data.data?.reqDTO?.filterWorkspaceId).toBe(wsId);
+  });
+
+  test("usage stats --dry-run --days 30 时间跨度约 30 天", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--dry-run",
+      "--days",
+      "30",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { reqDTO?: { startTime?: number; endTime?: number } };
+    }>(stdout);
+    const span = (data.data?.reqDTO?.endTime ?? 0) - (data.data?.reqDTO?.startTime ?? 0);
+    const thirtyDaysMs = 30 * 24 * 60 * 60 * 1000;
+    expect(span).toBeGreaterThan(thirtyDaysMs - 5000);
+    expect(span).toBeLessThan(thirtyDaysMs + 5000);
+  });
+
+  test("usage stats --dry-run --model 指定模型使用 list API", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--dry-run",
+      "--model",
+      "qwen-turbo",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      api?: string;
+      data?: { reqDTO?: { model?: string; filterWorkspaceId?: string } };
+    }>(stdout);
+    expect(data.api).toContain("listModelUsageStatisticData");
+    expect(data.data?.reqDTO?.model).toBe("qwen-turbo");
+    expect(data.data?.reqDTO?.filterWorkspaceId).toBe(wsId);
+  });
+
+  test("usage stats --dry-run --type Text 传递 obsModelType", async () => {
+    const { stdout, stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--dry-run",
+      "--type",
+      "Text",
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+    const data = parseStdoutJson<{
+      data?: { reqDTO?: { obsModelType?: string } };
+    }>(stdout);
+    expect(data.data?.reqDTO?.obsModelType).toBe("Text");
+  });
+
+  test("usage stats 概览模式返回 JSON 结果", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--output",
+      "json",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats 概览文本输出包含英文标签", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats 概览文本输出包含 Token 用量", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats --model 单模型文本输出包含英文表头", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--model",
+      "qwen3.6-plus",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats --model 逗号分隔多模型返回多行", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--model",
+      "qwen3.6-plus,deepseek-v4-pro",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats --model 不存在的模型返回空表格", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--model",
+      "nonexistent-model-xyz-99999",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats --days 1 短时间范围正常返回", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--days",
+      "1",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+
+  test("usage stats --type Vision 按类型过滤", async () => {
+    const { stderr, exitCode } = await runCli([
+      "usage",
+      "stats",
+      "--workspace-id",
+      wsId,
+      "--type",
+      "Vision",
+      "--output",
+      "text",
+      "--no-color",
+    ]);
+    expect(exitCode, stderr).toBe(0);
+  });
+});
@@ -91,7 +91,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "generate",
        "--model",
-        "happyhorse-1.0-t2v",
+        "happyhorse-1.1-t2v",
        "--duration",
        "3",
        "--prompt",
@@ -37,7 +37,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "generate",
        "--model",
-        "happyhorse-1.0-i2v",
+        "happyhorse-1.1-i2v",
        "--image",
        "https://example.com/placeholder.png",
        "--non-interactive",
@@ -53,7 +53,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "generate",
        "--dry-run",
        "--model",
-        "happyhorse-1.0-t2v",
+        "happyhorse-1.1-t2v",
        "--prompt",
        "干跑无图",
        "--non-interactive",
@@ -68,7 +68,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
      expect(data.request?.input?.media).toBeUndefined();
    });

-    test("【happyhorse-1.0-i2v】图片生成视频", async () => {
+    test("【happyhorse-1.1-i2v】图片生成视频", async () => {
      const outDir = makeE2eOutputDir(e2eLabelFromMetaUrl(import.meta.url));
      const png = join(outDir, "e2e-gen.png");
      const gen = await runCli([
@@ -95,7 +95,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "generate",
        "--model",
-        "happyhorse-1.0-i2v",
+        "happyhorse-1.1-i2v",
        "--image",
        imagePath,
        "--prompt",
@@ -37,7 +37,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "generate",
        "--model",
-        "happyhorse-1.0-t2v",
+        "happyhorse-1.1-t2v",
        "--non-interactive",
      ]);
      expect(exitCode).toBe(0);
@@ -51,7 +51,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "generate",
        "--dry-run",
        "--model",
-        "happyhorse-1.0-t2v",
+        "happyhorse-1.1-t2v",
        "--prompt",
        "干跑校验",
        "--non-interactive",
@@ -62,18 +62,18 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
      const data = parseStdoutJson<{ request?: { model?: string; input?: { prompt?: string } } }>(
        stdout,
      );
-      expect(data.request?.model).toBe("happyhorse-1.0-t2v");
+      expect(data.request?.model).toBe("happyhorse-1.1-t2v");
      expect(data.request?.input?.prompt).toBe("干跑校验");
    });

-    test("【happyhorse-1.0-t2v】文本生成视频", async () => {
+    test("【happyhorse-1.1-t2v】文本生成视频", async () => {
      const outDir = makeE2eOutputDir(e2eLabelFromMetaUrl(import.meta.url));
      const { stdout, stderr, exitCode } = await runCli([
        ...cliTimeoutPrefix(),
        "video",
        "generate",
        "--model",
-        "happyhorse-1.0-t2v",
+        "happyhorse-1.1-t2v",
        "--prompt",
        "夕阳下海面波光，远景静态镜头",
        "--download",
@@ -37,7 +37,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "ref",
        "--model",
-        "happyhorse-1.0-r2v",
+        "happyhorse-1.1-r2v",
        "--image",
        "https://example.com/x.png",
        "--non-interactive",
@@ -52,7 +52,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "ref",
        "--model",
-        "happyhorse-1.0-r2v",
+        "happyhorse-1.1-r2v",
        "--prompt",
        "仅有描述无素材",
        "--non-interactive",
@@ -61,7 +61,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
      expect(stderr).toMatch(/--image|ref-video|At least one|required/i);
    });

-    test("【happyhorse-1.0-r2v】视频参考生成", async () => {
+    test("【happyhorse-1.1-r2v】视频参考生成", async () => {
      const outDir = makeE2eOutputDir(e2eLabelFromMetaUrl(import.meta.url));
      const gen = await runCli([
        "image",
@@ -69,7 +69,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "--model",
        "qwen-image-2.0",
        "--prompt",
-        "一只简笔画小猫，白底",
+        "一片绿色的树叶，白底",
        "--out-dir",
        outDir,
        "--out-prefix",
@@ -88,7 +88,7 @@ describe.skipIf(!isBailianE2EVideoEnabled() || !isDashScopeE2EReady())(
        "video",
        "ref",
        "--model",
-        "happyhorse-1.0-r2v",
+        "happyhorse-1.1-r2v",
        "--prompt",
        "图1在画面中心轻微晃动",
        "--image",
@@ -0,0 +1,42 @@
+import { expect, test } from "vite-plus/test";
+import { readProxyEnv } from "../src/proxy.ts";
+
+test("readProxyEnv: 未设置任何代理变量时全部为 undefined", () => {
+  expect(readProxyEnv({})).toEqual({
+    httpProxy: undefined,
+    httpsProxy: undefined,
+    noProxy: undefined,
+  });
+});
+
+test("readProxyEnv: 空白值视为未设置", () => {
+  expect(readProxyEnv({ HTTPS_PROXY: "", HTTP_PROXY: "   ", NO_PROXY: "" })).toEqual({
+    httpProxy: undefined,
+    httpsProxy: undefined,
+    noProxy: undefined,
+  });
+});
+
+test("readProxyEnv: 大小写变量均可识别，小写优先", () => {
+  expect(readProxyEnv({ HTTPS_PROXY: "http://upper:1" }).httpsProxy).toBe("http://upper:1");
+  expect(readProxyEnv({ https_proxy: "http://lower:1" }).httpsProxy).toBe("http://lower:1");
+  expect(
+    readProxyEnv({ https_proxy: "http://lower:1", HTTPS_PROXY: "http://upper:1" }).httpsProxy,
+  ).toBe("http://lower:1");
+});
+
+test("readProxyEnv: 空字符串小写变量不屏蔽已设置的大写变量", () => {
+  expect(readProxyEnv({ https_proxy: "", HTTPS_PROXY: "http://upper:1" }).httpsProxy).toBe(
+    "http://upper:1",
+  );
+  expect(readProxyEnv({ http_proxy: "", HTTP_PROXY: "http://upper:2" }).httpProxy).toBe(
+    "http://upper:2",
+  );
+});
+
+test("readProxyEnv: NO_PROXY 独立读取", () => {
+  const r = readProxyEnv({ NO_PROXY: "*.aliyuncs.com" });
+  expect(r.noProxy).toBe("*.aliyuncs.com");
+  expect(r.httpProxy).toBeUndefined();
+  expect(r.httpsProxy).toBeUndefined();
+});
@@ -180,7 +180,7 @@ export async function ensurePrerequisites(ctx) {
      "video",
      "generate",
      "--model",
-      "happyhorse-1.0-t2v",
+      "happyhorse-1.1-t2v",
      "--prompt",
      "压测前置短视频：海浪与静态远景，无明显人物。",
      "--duration",
@@ -132,7 +132,7 @@ export async function generateCombinedFixtures({ suiteRoot, cliPackage }) {
      "video",
      "generate",
      "--model",
-      "happyhorse-1.0-t2v",
+      "happyhorse-1.1-t2v",
      "--prompt",
      "压测前置短视频：海浪与静态远景，无明显人物。",
      "--duration",
@@ -16,7 +16,7 @@ const motions = [

 export const runStress = defineStressTarget({
  canonical: "video-i2v",
-  defaultModel: "happyhorse-1.0-i2v",
+  defaultModel: "happyhorse-1.1-i2v",
  batchDirPrefix: "video-i2v-batch",
  helpText: "pnpm run test:stress -- video-i2v [--reuse-fixtures] -- --count 5 -c 2",

@@ -16,7 +16,7 @@ const prompts = [

 export const runStress = defineStressTarget({
  canonical: "video-ref",
-  defaultModel: "happyhorse-1.0-r2v",
+  defaultModel: "happyhorse-1.1-r2v",
  batchDirPrefix: "video-ref-batch",
  helpText: "pnpm run test:stress -- video-ref [--reuse-fixtures] -- --count 5 -c 2",

@@ -45,7 +45,7 @@ const pick = (arr) => arr[Math.floor(Math.random() * arr.length)];

 export const runStress = defineStressTarget({
  canonical: "video-t2v",
-  defaultModel: "happyhorse-1.0-t2v",
+  defaultModel: "happyhorse-1.1-t2v",
  batchDirPrefix: "video-t2v-batch",
  helpText: `用法：pnpm run test:stress -- video-t2v -- --concurrency 1 --count 3
 详见 docs/agents/stress-batch-tests.md`,
@@ -1,6 +1,6 @@
 {
  "name": "bailian-cli-core",
-  "version": "1.2.1",
+  "version": "1.4.2",
  "description": "Core SDK for bailian-cli. See https://www.npmjs.com/package/bailian-cli for usage.",
  "homepage": "https://bailian.console.aliyun.com/cli",
  "bugs": {
@@ -1,196 +1,181 @@
-export const INTENT_MODEL = "qwen-turbo";
+export const INTENT_MODEL = "qwen-flash";
 export const RANKING_MODEL = "qwen3.6-flash";
-export const RANKING_MODEL_FAST = "qwen-turbo";
+export const RANKING_MODEL_FAST = "qwen-flash";

-export const INTENT_SYSTEM_PROMPT = `你是一个意图分析器。根据用户的需求描述，先理解用户场景，再提取结构化信息。
+export const INTENT_SYSTEM_PROMPT = `You are an intent analyzer. Given the user's requirement, understand the scenario first, then extract structured information.

-## 分析步骤
-1. 用一句话总结用户的核心需求（taskSummary），要体现具体场景而非泛泛描述
-2. 推断场景特征（scenarioHints），例如：["需要低延迟","面向C端用户","高并发","对话式交互","离线批处理","需要精准度"]
-3. 基于场景特征推断 budget 和 qualityPreference
-   - 只在用户明确表达或场景强烈暗示时偏离默认值
-   - 用户明确说"低成本"、"便宜"、"省钱" → budget:"low"
-   - 用户明确说"最好的"、"高精度"、"不计成本" → qualityPreference:"flagship"
-   - 场景本身有强约束时才推断：如"日均百万请求的客服" → budget:"low"（高并发=成本敏感）
-   - 其他情况保持 budget:"medium", qualityPreference:"balanced"
-4. 提取模态、能力、特性等结构化字段
+CRITICAL: You MUST respond entirely in English. Do not use any Chinese characters anywhere in your response. All text fields (taskSummary, scenarioHints) must be in English.

-## 示例
+## Analysis Steps
+1. Summarize the user's core need in one sentence (taskSummary) — be specific about the scenario, not generic
+2. Infer scenario hints (scenarioHints), e.g.: ["low-latency", "consumer-facing", "high-concurrency", "conversational", "offline-batch", "high-precision"]
+3. Infer budget and qualityPreference from scenario hints
+   - Only deviate from defaults when the user explicitly states or the scenario strongly implies
+   - User says "low cost", "cheap", "save money" → budget:"low"
+   - User says "best", "high precision", "cost no object" → qualityPreference:"flagship"
+   - Infer from scenario constraints only when strong: e.g. "1M requests/day customer service" → budget:"low" (high concurrency = cost-sensitive)
+   - Otherwise keep budget:"medium", qualityPreference:"balanced"
+4. Extract modalities, capabilities, features etc.

-用户: "做一个低成本高并发的在线客服"
-→ budget:"low", qualityPreference:"cost-optimized"（用户明确说了低成本）
+## Model preference detection
+Analyze whether the user mentioned specific models, model families, or vendors:
+- No models/families/vendors mentioned → mode:"unconstrained", no targets
+- User scoped the range (e.g. "recommend from the deepseek family", "open-source reasoning models") → mode:"scoped", targets:["deepseek"]
+- User wants to compare specific models (e.g. "compare wan2.6 and wan2.7", "is qwen-max good for legal analysis") → mode:"comparison", targets:["wan2.6","wan2.7"]
+  - Single model evaluation is also comparison with one target
+- User wants alternatives to a reference model (e.g. "something like qwen-max but cheaper") → mode:"alternative", targets:["qwen-max"]
+- User explicitly excludes certain models/families (e.g. "good models besides qwen") → excludes:["qwen"], mode determined by other signals
+- targets should capture the model/family names as the user wrote them

-用户: "法律合同审查，要求高精准度"
-→ budget:"medium", qualityPreference:"flagship"（用户明确要求高精准度，但没提预算）
-
-用户: "我要做一个能理解图片的客服机器人"
-→ budget:"medium", qualityPreference:"balanced"（用户没提成本和质量要求，不过度推断）
-
-用户: "帮我选一个写代码的模型"
-→ budget:"medium", qualityPreference:"balanced"（通用需求，无明确倾向）
-
-用户: "预算有限，做个简单的文本摘要功能"
-→ budget:"low", qualityPreference:"cost-optimized"（用户说了预算有限）
-
-用户: "企业级知识库问答，准确率是第一优先级"
-→ budget:"high", qualityPreference:"flagship"（企业级+准确率第一=愿投入高成本）
-
-用户: "个人学习项目，试试AI生成图片"
-→ budget:"low", qualityPreference:"cost-optimized"（个人学习=成本敏感）
-
-用户: "做一个Agent自动根据用户意图生成动画片"
-→ budget:"medium", qualityPreference:"balanced"（复杂pipeline，但没明确成本/质量约束）
-
-## 模型偏好识别
-分析用户是否提到了特定的模型、模型系列或厂商，据此判断推荐模式：
- 用户未提到任何模型/系列/厂商 → mode:"unconstrained"，不填 targets
- 用户限定了范围（如"deepseek系列哪个好"、"通义千问的模型推荐"、"开源的推理模型"） → mode:"scoped"，targets:["deepseek"] 或 ["通义千问"]
- 用户要对比特定模型（如"wan2.6和wan2.7哪个好"、"qwen-max和deepseek-v3对比"、"qwen-max适合做法律分析吗"） → mode:"comparison"，targets:["wan2.6","wan2.7"]
-  - 单模型评估也算 comparison，targets 只填一个
- 用户以某模型为参照找替代（如"有没有类似qwen-max但更便宜的"） → mode:"alternative"，targets:["qwen-max"]
- 用户明确排除某些模型/系列（如"除了qwen还有什么好的"） → excludes:["qwen"]，mode 根据其他条件判断
- targets 填写用户原文中的模型/系列名称，保持原文写法
-
-## 输出字段
- taskSummary: 一句话场景理解（必须具体，禁止"用户想用AI做某事"这种废话）
- scenarioHints: 推断的场景特征数组
- complexity: "single"（单一模型可完成）或 "pipeline"（需要多个模型协同）
- segments: 仅 pipeline 时填写，每步包含 step/inputModality/outputModality/requiredCapabilities。
-  - step 必须是一句话描述该步骤在用户任务中解决的具体问题，例如"解析天气预报数据，生成适合视频制作的场景描述文本"，禁止用编号或泛化的模态标签
-  - segments 必须形成模态链路：每步的 inputModality 应包含上一步的 outputModality，确保上下游数据可以衔接
- inputModality: 用户输入涉及的模态 ["Text","Image","Video","Audio"]
- outputModality: 期望输出的模态
- requiredCapabilities: 需要的能力。可选代码（必须严格使用，不要自创）：
-  TG=文本生成, Reasoning=推理, VU=视觉理解, IG=图像生成, VG=视频生成,
-  TTS=语音合成, ASR=语音识别, Realtime-ASR=实时语音识别,
-  Realtime-Text-to-Speech=实时语音合成, Realtime-Audio-Translate=实时音频翻译,
-  Realtime-Omni=实时全模态, Multimodal-Omni=全模态, ME=多模态嵌入,
-  TR=翻译, 3D-generation=3D生成
- requiredFeatures: 需要的特性 (function-calling, web-search, structured-outputs, prefix-completion)
- budget: "low"/"medium"/"high"（基于场景推断，不要默认 medium）
+## Output fields
+- taskSummary: one-sentence scenario understanding (must be specific, never generic like "user wants AI")
+- scenarioHints: array of inferred scenario features
+- complexity: "single" or "pipeline"
+- segments: only for pipeline, each with step/inputModality/outputModality/requiredCapabilities
+  - step must describe the specific problem this step solves in the user's task, no numbered or generic modal labels
+  - segments must form a modality chain: each step's inputModality should cover the previous step's outputModality
+- inputModality: user input modalities ["Text","Image","Video","Audio"]
+- outputModality: expected output modalities
+- requiredCapabilities: capability codes (use strictly from the list, don't invent):
+  TG=Text Generation, Reasoning=Reasoning, VU=Vision Understanding, IG=Image Generation, VG=Video Generation,
+  TTS=Text-to-Speech, ASR=Speech-to-Text, Realtime-ASR=Realtime Speech-to-Text,
+  Realtime-Text-to-Speech=Realtime Text-to-Speech, Realtime-Audio-Translate=Realtime Audio Translation,
+  Realtime-Omni=Realtime Omni-modal, Multimodal-Omni=Multimodal Omni, ME=Multimodal Embedding,
+  TR=Translation, 3D-generation=3D Generation
+- requiredFeatures: required features (function-calling, web-search, structured-outputs, prefix-completion)
+- budget: "low"/"medium"/"high"
 - contextNeed: "standard"/"large"/"extra-large"
- qualityPreference: "flagship"/"balanced"/"cost-optimized"（基于场景推断，不要默认 balanced）
- modelPreference: { mode, targets?, excludes? }（见上方"模型偏好识别"）
+- qualityPreference: "flagship"/"balanced"/"cost-optimized"
+- modelPreference: { mode, targets?, excludes? }

-只输出 JSON，不要有其他文字。`;
+Output only JSON, no other text.`;

-export const SINGLE_SYSTEM_PROMPT = `你是阿里云百炼平台的模型推荐顾问。从以下候选模型中选出最佳推荐。
+export const SINGLE_SYSTEM_PROMPT = `You are a model recommendation advisor for Alibaba Cloud Model Studio. From the candidate models below, select the best recommendations.

-## 背景
-系统已根据用户意图预筛选了候选模型，你只需从中精选并排序。
-意图分析中包含 budget 和 qualityPreference 字段，这代表了用户的实际需求层次。
+CRITICAL: You MUST respond entirely in English. Do not use any Chinese characters anywhere in your response. Every field — reason, highlights, step, summary — must be written in English.

-## 推荐策略
+## Background
+The system has pre-filtered candidate models based on intent analysis. Your job is to rank and pick from these candidates.
+The intent includes budget and qualityPreference fields representing the user's actual needs.

-推荐 3 个不同档次的模型，但排序必须反映用户的真实需求：
+## Recommendation Strategy

- 推荐 #1（最佳推荐）：根据 budget 和 qualityPreference 判断哪个档次最适合用户，把那个档次的最佳模型放在第一位
- 推荐 #2（次优选择）：另一个档次中值得考虑的模型，说明与 #1 相比的 tradeoff
- 推荐 #3（备选参考）：第三个视角的选择，说明适用场景差异
+Recommend 3 models at different tiers, but ordering must reflect the user's true needs:

-关键原则：
- budget:"low" / qualityPreference:"cost-optimized" → 推荐 #1 应该是性价比最高的模型，而非旗舰模型
- budget:"high" / qualityPreference:"flagship" → 推荐 #1 应该是能力最强的旗舰模型
- budget:"medium" / qualityPreference:"balanced" → 推荐 #1 应该是综合匹配度最高的模型，不预设档次偏好
+- #1 (Best Pick): Based on budget and qualityPreference, pick the best-fitting tier and put its top model first
+- #2 (Runner-Up): A worthy consideration from another tier, explaining tradeoffs vs #1
+- #3 (Alternative): A third-perspective choice, explaining scenario differences

-每个推荐都必须说明该模型为什么适合（或作为备选为什么值得考虑），理由必须关联用户的具体需求。
+Key principles:
+- budget:"low" / qualityPreference:"cost-optimized" → #1 should be the best value model, not a flagship
+- budget:"high" / qualityPreference:"flagship" → #1 should be the most capable flagship model
+- budget:"medium" / qualityPreference:"balanced" → #1 should be the best all-around match

-## 规则
- 只能推荐候选列表中的模型，严禁推荐列表外的模型
- 严禁使用泛泛的推荐理由（如"性能强大"、"综合能力好"、"效果不错"），每条 reason 必须说明该模型解决用户任务中的什么具体问题
- 三个推荐的理由不允许雷同，每个必须从不同维度论证
- 有定价信息时：结合 budget 字段权衡，把最符合用户预算的放在最前面
- 有家族信息时：避免推荐同一家族的多个模型，优先推荐稳定版本
- 有版本标签时：优先推荐 stable/latest 版本，除非用户明确需要特定版本
- 没有增强字段的模型：按能力和描述排序即可，不因缺少信息而降权
- 如果没有合适的模型，返回空数组
- 如果你认为该需求实际需要多模型协同完成（pipeline），可以输出 type:"pipeline" 格式
- 输出严格 JSON，不要输出其他内容
+Each recommendation must explain why the model fits (or as an alternative, why it's worth considering), with reasoning tied to the user's specific needs.

-## 输出格式
+## Rules
+- Only recommend models from the candidate list — never recommend outside it
+- No generic reasons ("powerful", "good performance", "effective"). Each reason must describe how the model solves a specific aspect of the user's task
+- All three recommendations must have distinct reasoning angles, not duplicate reasons
+- When pricing is available: factor in budget, put the most budget-friendly option first
+- When family info is available: avoid recommending multiple models from the same family, prefer stable versions
+- When version tags are available: prefer stable/latest versions unless the user explicitly needs a specific version
+- Models without enriched fields: rank by capability and description — don't penalize for missing info
+- If no model fits, return an empty array
+- If you believe the task actually requires multi-model collaboration (pipeline), you may output type:"pipeline" format
+- Output strict JSON, no other text

-单一任务：
-{"type":"single","recommendations":[{"model":"模型ID","reason":"推荐理由","highlights":["亮点"]}]}
+## Output Format

-复合任务（仅当你确信需要多模型协同时）：
-{"type":"pipeline","summary":"一句话方案描述","steps":[{"step":"步骤描述","recommendations":[{"model":"模型ID","reason":"选择理由","highlights":["亮点"]}]}]}`;
+Single task:
+{"type":"single","recommendations":[{"model":"model ID","reason":"recommendation reason","highlights":["key highlights"]}]}

-export const PIPELINE_SYSTEM_PROMPT = `你是阿里云百炼平台的模型推荐顾问。用户需求已被拆解为多步骤流水线，请为每步选出最佳模型。
+Pipeline (only when confident multi-model is needed):
+{"type":"pipeline","summary":"one-line solution description","steps":[{"step":"step description","recommendations":[{"model":"model ID","reason":"reason for choosing","highlights":["highlights"]}]}]}`;

-## 背景
-系统已根据各步骤需求预筛选了候选模型。
-意图分析中包含 budget 和 qualityPreference 字段，这代表了用户的实际需求层次。
+export const PIPELINE_SYSTEM_PROMPT = `You are a model recommendation advisor for Alibaba Cloud Model Studio. The user's need has been decomposed into multi-step pipeline. Select the best model for each step.

-## 推荐策略
+CRITICAL: You MUST respond entirely in English. Do not use any Chinese characters anywhere in your response. Every field — reason, highlights, step, summary — must be written in English.

-每步推荐 3 个不同档次的模型，但排序必须反映用户的真实需求：
+## Background
+The system has pre-filtered candidate models for each step's requirements.
+The intent includes budget and qualityPreference fields representing the user's actual needs.

- 推荐 #1（最佳推荐）：根据 budget 和 qualityPreference 判断哪个档次最适合用户，把那个档次的最佳模型放在第一位
- 推荐 #2（次优选择）：另一个档次中值得考虑的模型，说明 tradeoff
- 推荐 #3（备选参考）：第三个视角的选择，说明适用场景差异
+## Recommendation Strategy

-关键原则：
- budget:"low" / qualityPreference:"cost-optimized" → 推荐 #1 应该是性价比最高的模型
- budget:"high" / qualityPreference:"flagship" → 推荐 #1 应该是能力最强的旗舰模型
- budget:"medium" / qualityPreference:"balanced" → 推荐 #1 应该是综合匹配度最高的模型
+Recommend 3 models at different tiers per step, ordering by user needs:

-## 规则
- 只能推荐候选列表中的模型
- 每步推荐多个模型，按优先级排序，每个推荐给出简短理由和关键亮点
- step 字段必须用一句话描述该步骤在用户任务中解决的具体问题，禁止用编号或泛化的模态标签（如"输出: Text"）
- 严禁使用泛泛的推荐理由，每条 reason 必须说明该模型在这一步解决用户任务中的什么具体问题
- 有定价信息时：结合 budget 字段权衡，把最符合用户预算的放在最前面
- 有家族信息时：避免在相邻步骤使用同一家族的不同规格模型，除非确实需要
- 没有增强字段的模型：按能力和描述排序即可，不因缺少信息而降权
- 相邻步骤的模型必须模态兼容：上一步模型的输出模态必须被下一步模型的输入模态支持
- 如果你认为该需求其实单模型可以完成，可以输出 type:"single" 格式
- 输出严格 JSON
+- #1 (Best Pick): Based on budget and qualityPreference, pick the best-fitting tier and put its top model first
+- #2 (Runner-Up): A worthy consideration from another tier, explaining tradeoffs
+- #3 (Alternative): A third-perspective choice

-## 输出格式
+Key principles:
+- budget:"low" / qualityPreference:"cost-optimized" → #1 should be the best value model
+- budget:"high" / qualityPreference:"flagship" → #1 should be the most capable flagship model
+- budget:"medium" / qualityPreference:"balanced" → #1 should be the best all-around match

-{"type":"pipeline","summary":"一句话方案描述","steps":[{"step":"该步骤在用户任务中解决的具体问题","recommendations":[{"model":"模型ID","reason":"该模型如何解决这一步的具体问题","highlights":["亮点"]}]}]}
+## Rules
+- Only recommend models from the candidate list
+- Each step recommends multiple models sorted by priority, each with brief reason and key highlights
+- The "step" field must describe the specific problem this step solves in the user's task — no numbered or generic modal labels (e.g. "Output: Text")
+- No generic reasons. Each reason must describe how the model solves a specific aspect of the user's task at this step
+- When pricing is available: factor in budget, put the most budget-friendly option first
+- When family info is available: avoid using different tiers of the same family in adjacent steps unless truly needed
+- Models without enriched fields: rank by capability and description — don't penalize for missing info
+- Adjacent steps must be modality-compatible: the previous step's output modalities must be supported as input modalities by the next step
+- If you believe the task can be done with a single model, output type:"single" format
+- Output strict JSON

-或者（如果你认为单模型即可）：
-{"type":"single","recommendations":[{"model":"模型ID","reason":"推荐理由","highlights":["亮点"]}]}`;
+## Output Format

-export const COMPARISON_SYSTEM_PROMPT = `你是阿里云百炼平台的模型对比顾问。用户想对比特定模型，请根据使用场景进行对比分析。
+{"type":"pipeline","summary":"one-line solution description","steps":[{"step":"specific problem this step solves in the user's task","recommendations":[{"model":"model ID","reason":"how this model solves the specific problem at this step","highlights":["highlights"]}]}]}

-## 背景
-用户指定了要对比的模型，系统已将这些模型和相关候选预筛选到列表中。
-意图分析中的 modelPreference.targets 是用户要对比的模型。
+Or (if single model suffices):
+{"type":"single","recommendations":[{"model":"model ID","reason":"recommendation reason","highlights":
+["key highlights"]}]}`;

-## 对比策略
- 用户指定的模型必须全部出现在推荐结果中，按适合程度排序
- 每个模型的 reason 必须是对比性的，说明该模型相对于其他对比模型的优势和劣势
- 如果候选中有比用户指定的更合适的模型，可以额外推荐，但用户指定的必须优先包含
- 单模型评估场景（targets 只有一个）：评估该模型是否适合用户需求，同时推荐更优的替代
+export const COMPARISON_SYSTEM_PROMPT = `You are a model comparison advisor for Alibaba Cloud Model Studio. The user wants to compare specific models — analyze them against the use case.

-## 规则
- 只能推荐候选列表中的模型
- reason 必须包含对比视角：该模型相比其他模型在哪些方面更好/更差
- highlights 突出各模型的差异化特点
- 输出严格 JSON，不要输出其他内容
+CRITICAL: You MUST respond entirely in English. Do not use any Chinese characters anywhere in your response. Every field — reason, highlights — must be written in English.

-## 输出格式
-{"type":"single","recommendations":[{"model":"模型ID","reason":"对比分析理由","highlights":["差异化亮点"]}]}`;
+## Background
+The user specified models to compare. The system has pre-filtered these models and related candidates into the list.
+The intent's modelPreference.targets are the models to compare.

-export const ALTERNATIVE_SYSTEM_PROMPT = `你是阿里云百炼平台的模型替代顾问。用户以某个模型为参照，寻找替代方案。
+## Comparison Strategy
+- All user-specified models must appear in the results, sorted by suitability
+- Each model's reason must be comparative: describe strengths and weaknesses relative to other models being compared
+- If candidates contain better fits than what the user specified, they can be additionally recommended, but user-specified models take priority
+- Single-model evaluation (one target): evaluate if the model fits, and recommend better alternatives

-## 背景
-用户以某个模型为参照点，想找到在特定维度上更优的替代方案（如更便宜、更快、更强）。
-意图分析中的 modelPreference.targets 是参照模型。
+## Rules
+- Only recommend models from the candidate list
+- reason must include comparative perspective: how this model is better/worse compared to others
+- highlights should emphasize differentiating characteristics
+- Output strict JSON

-## 替代策略
- 推荐 #1：如果参照模型在候选中，先评估它是否满足用户需求，给出其基本定位
- 推荐 #2~#3：推荐替代方案，reason 必须说明相比参照模型在用户关注维度上的 tradeoff
- 关注用户提到的替代维度（如"更便宜"→重点对比定价，"更强"→重点对比能力）
+## Output Format
+{"type":"single","recommendations":[{"model":"model ID","reason":"comparative analysis","highlights":["differentiators"]}]}`;

-## 规则
- 只能推荐候选列表中的模型
- 参照模型必须包含在结果中（如果在候选列表中）
- 替代推荐的 reason 必须说明与参照模型的具体差异
- 避免推荐和参照模型同系列的其他版本（除非确实有显著差异）
- 输出严格 JSON，不要输出其他内容
+export const ALTERNATIVE_SYSTEM_PROMPT = `You are a model alternative advisor for Alibaba Cloud Model Studio. The user has a reference model and wants to find alternatives.

-## 输出格式
-{"type":"single","recommendations":[{"model":"模型ID","reason":"替代分析理由","highlights":["差异化亮点"]}]}`;
+CRITICAL: You MUST respond entirely in English. Do not use any Chinese characters anywhere in your response. Every field — reason, highlights — must be written in English.
+
+## Background
+The user has a reference model and wants to find alternatives that are better in specific dimensions (cheaper, faster, more capable).
+The intent's modelPreference.targets is the reference model.
+
+## Alternative Strategy
+- #1: If the reference model is in candidates, first evaluate if it meets the user's needs — give its positioning
+- #2~#3: Recommend alternatives. reason must explain the tradeoff vs the reference model in the user's dimensions of interest
+- Focus on the user's stated alternative dimension (e.g. "cheaper" → focus on pricing comparison, "better" → focus on capability comparison)
+
+## Rules
+- Only recommend models from the candidate list
+- The reference model must be included in results if it's in the candidate list
+- Alternative recommendations must explain concrete differences from the reference model
+- Avoid recommending other versions from the same family unless there's a significant difference
+- Output strict JSON
+
+## Output Format
+{"type":"single","recommendations":[{"model":"model ID","reason":"alternative analysis","highlights":["differentiators"]}]}`;
@@ -76,20 +76,20 @@ async function embedBatch(config: Config, texts: string[]): Promise<number[][]>
 }

 const CAPABILITY_LABELS: Record<string, string> = {
-  TG: "文本生成",
-  Reasoning: "推理",
-  VU: "视觉理解",
-  IG: "图像生成",
-  VG: "视频生成",
-  TTS: "语音合成",
-  ASR: "语音识别",
+  TG: "Text Generation",
+  Reasoning: "Reasoning",
+  VU: "Vision Understanding",
+  IG: "Image Generation",
+  VG: "Video Generation",
+  TTS: "Text-to-Speech",
+  ASR: "Speech-to-Text",
 };

 const MODALITY_LABELS: Record<string, string> = {
-  Text: "文本",
-  Image: "图片/图像",
-  Video: "视频",
-  Audio: "音频/语音",
+  Text: "Text",
+  Image: "Image",
+  Video: "Video",
+  Audio: "Audio",
 };

 interface GroupData {
@@ -135,12 +135,12 @@ function buildModelText(model: ModelProfile, descriptions: Map<string, string>):
    model.name,
    model.model,
    description,
-    caps ? `能力: ${caps}` : "",
-    inputMods ? `输入: ${inputMods}` : "",
-    outputMods ? `输出: ${outputMods}` : "",
-    model.features?.length ? `特性: ${model.features.join(", ")}` : "",
+    caps ? `Capabilities: ${caps}` : "",
+    inputMods ? `Input: ${inputMods}` : "",
+    outputMods ? `Output: ${outputMods}` : "",
+    model.features?.length ? `Features: ${model.features.join(", ")}` : "",
    model.familyName || "",
-    model.category ? `定位: ${model.category}` : "",
+    model.category ? `Category: ${model.category}` : "",
  ].filter(Boolean);

  return parts.join(" | ");
@@ -46,26 +46,27 @@ function buildCandidatesContext(candidates: ScoredCandidate[]): string {
    .map(({ model: profile }) => {
      const parts = [
        `ID: ${profile.model}`,
-        `名称: ${profile.name}`,
-        `描述: ${profile.shortDescription || profile.description}`,
-        `能力: ${profile.capabilities.join(", ")}`,
-        `特性: ${profile.features.join(", ")}`,
+        `Name: ${profile.name}`,
+        `Description: ${profile.shortDescription || profile.description}`,
+        `Capabilities: ${profile.capabilities.join(", ")}`,
+        `Features: ${profile.features.join(", ")}`,
      ];
-      if (profile.contextWindow) parts.push(`上下文窗口: ${profile.contextWindow}`);
-      if (profile.maxOutputTokens) parts.push(`最大输出: ${profile.maxOutputTokens}`);
-      if (profile.category) parts.push(`类别: ${profile.category}`);
+      if (profile.contextWindow) parts.push(`Context Window: ${profile.contextWindow}`);
+      if (profile.maxOutputTokens) parts.push(`Max Output: ${profile.maxOutputTokens}`);
+      if (profile.category) parts.push(`Category: ${profile.category}`);
      const modality = profile.inferenceMetadata;
      if (modality?.request_modality?.length)
-        parts.push(`输入模态: ${modality.request_modality.join(", ")}`);
+        parts.push(`Input Modality: ${modality.request_modality.join(", ")}`);
      if (modality?.response_modality?.length)
-        parts.push(`输出模态: ${modality.response_modality.join(", ")}`);
+        parts.push(`Output Modality: ${modality.response_modality.join(", ")}`);
      const prices = formatPrices(profile);
-      if (prices) parts.push(`定价: ${prices}`);
+      if (prices) parts.push(`Pricing: ${prices}`);
      const qpm = formatQpm(profile);
      if (qpm) parts.push(`QPM: ${qpm}`);
-      if (profile.versionTag) parts.push(`版本: ${profile.versionTag}`);
-      if (profile.openSource !== undefined) parts.push(`开源: ${profile.openSource ? "是" : "否"}`);
-      if (profile.family) parts.push(`家族: ${profile.family}`);
+      if (profile.versionTag) parts.push(`Version: ${profile.versionTag}`);
+      if (profile.openSource !== undefined)
+        parts.push(`Open Source: ${profile.openSource ? "Yes" : "No"}`);
+      if (profile.family) parts.push(`Family: ${profile.family}`);
      return parts.join(" | ");
    })
    .join("\n");
@@ -86,29 +87,29 @@ function buildIntentContext(intent: IntentProfile): string {
    modelPreference,
  } = intent;
  const parts: string[] = [];
-  if (taskSummary) parts.push(`场景理解: ${taskSummary}`);
-  if (scenarioHints.length) parts.push(`场景特征: ${scenarioHints.join(", ")}`);
-  if (inputModality.length) parts.push(`输入模态: ${inputModality.join(", ")}`);
-  if (outputModality.length) parts.push(`输出模态: ${outputModality.join(", ")}`);
-  if (requiredCapabilities.length) parts.push(`所需能力: ${requiredCapabilities.join(", ")}`);
-  if (requiredFeatures.length) parts.push(`所需特性: ${requiredFeatures.join(", ")}`);
-  parts.push(`预算倾向: ${budget}`);
-  parts.push(`质量偏好: ${qualityPreference}`);
-  if (contextNeed !== ContextNeeds.Standard) parts.push(`上下文需求: ${contextNeed}`);
+  if (taskSummary) parts.push(`Task: ${taskSummary}`);
+  if (scenarioHints.length) parts.push(`Scenario: ${scenarioHints.join(", ")}`);
+  if (inputModality.length) parts.push(`Input: ${inputModality.join(", ")}`);
+  if (outputModality.length) parts.push(`Output: ${outputModality.join(", ")}`);
+  if (requiredCapabilities.length) parts.push(`Capabilities: ${requiredCapabilities.join(", ")}`);
+  if (requiredFeatures.length) parts.push(`Features: ${requiredFeatures.join(", ")}`);
+  parts.push(`Budget: ${budget}`);
+  parts.push(`Quality: ${qualityPreference}`);
+  if (contextNeed !== ContextNeeds.Standard) parts.push(`Context: ${contextNeed}`);
  if (modelPreference && modelPreference.mode !== "unconstrained") {
-    parts.push(`模型偏好: ${modelPreference.mode}`);
+    parts.push(`Mode: ${modelPreference.mode}`);
    if (modelPreference.targets?.length)
-      parts.push(`目标模型: ${modelPreference.targets.join(", ")}`);
+      parts.push(`Targets: ${modelPreference.targets.join(", ")}`);
    if (modelPreference.excludes?.length)
-      parts.push(`排除模型: ${modelPreference.excludes.join(", ")}`);
+      parts.push(`Excludes: ${modelPreference.excludes.join(", ")}`);
  }
  if (segments?.length) {
-    parts.push(`拆解步骤:`);
+    parts.push(`Pipeline Steps:`);
    for (const seg of segments) {
-      const inMod = seg.inputModality.join(",") || "无";
-      const outMod = seg.outputModality.join(",") || "无";
-      const caps = seg.requiredCapabilities.join(",") || "无";
-      parts.push(`  - ${seg.step} (输入: ${inMod} → 输出: ${outMod}, 能力: ${caps})`);
+      const inMod = seg.inputModality.join(",") || "none";
+      const outMod = seg.outputModality.join(",") || "none";
+      const caps = seg.requiredCapabilities.join(",") || "none";
+      parts.push(`  - ${seg.step} (Input: ${inMod} → Output: ${outMod}, Capabilities: ${caps})`);
    }
  }
  return parts.join("\n");
@@ -175,7 +176,7 @@ function validatePipelineCompatibility(
      const compatible = accepts.some((mod) => prevOutputs.has(mod));
      if (!compatible && accepts.length > 0) {
        warnings.push(
-          `${rec.name} 的输入模态 [${accepts.join(", ")}] 可能不兼容上一步的输出模态 [${[...prevOutputs].join(", ")}]`,
+          `${rec.name}'s input modalities [${accepts.join(", ")}] may not be compatible with the previous step's output modalities [${[...prevOutputs].join(", ")}]`,
        );
      }
    }
@@ -204,7 +205,7 @@ export async function rankModels(
    systemPrompt = ALTERNATIVE_SYSTEM_PROMPT;
  } else if (preferenceMode === "scoped") {
    const scopeNote = intent.modelPreference?.targets?.length
-      ? `\n\n## 范围限定\n用户明确要求在以下范围内推荐：${intent.modelPreference.targets.join("、")}。请优先从匹配该范围的模型中选择。`
+      ? `\n\n## Scope Restriction\nThe user explicitly requested recommendations from: ${intent.modelPreference.targets.join(", ")}. Prioritize models within this scope.`
      : "";
    systemPrompt =
      (intent.complexity === Complexities.Pipeline
@@ -219,8 +220,8 @@ export async function rankModels(

  const userMessage =
    intent.complexity === Complexities.Pipeline
-      ? `意图分析结果：\n${intentContext}\n\n候选模型列表：\n${candidatesContext}\n\n用户原始需求：${userInput}\n\n请为流水线各步骤各推荐最多 ${top} 个模型。`
-      : `意图分析结果：\n${intentContext}\n\n候选模型列表：\n${candidatesContext}\n\n用户原始需求：${userInput}\n\n请推荐最多 ${top} 个模型。`;
+      ? `Intent Analysis:\n${intentContext}\n\nCandidate Models:\n${candidatesContext}\n\nUser Request: ${userInput}\n\nRecommend up to ${top} models for each pipeline step. Respond in English only.`
+      : `Intent Analysis:\n${intentContext}\n\nCandidate Models:\n${candidatesContext}\n\nUser Request: ${userInput}\n\nRecommend up to ${top} models. Respond in English only.`;

  const body: Record<string, unknown> = {
    model: useThinkingModel ? RANKING_MODEL : RANKING_MODEL_FAST,
@@ -73,6 +73,12 @@ export function userProfileEndpoint(baseUrl: string, schemaId: string): string {
  return `${baseUrl}/api/v2/apps/memory/profile_schemas/${encodeURIComponent(schemaId)}/profiles`;
 }

+// ---- Knowledge Base Retrieve (DashScope) ----
+
+export function knowledgeRetrieveEndpoint(baseUrl: string): string {
+  return `${baseUrl}/api/v1/indices/rag/index/retrieve`;
+}
+
 // ---- MCP Services (Streamable HTTP) ----

 export function mcpWebSearchEndpoint(baseUrl: string): string {
@@ -5,6 +5,7 @@ export {
  chatEndpoint,
  imageEndpoint,
  imageSyncEndpoint,
+  knowledgeRetrieveEndpoint,
  memoryAddEndpoint,
  memoryListEndpoint,
  memoryNodeEndpoint,
@@ -1,5 +1,5 @@
 import { readFileSync, writeFileSync, renameSync, existsSync } from "fs";
-import { parseConfigFile, REGIONS, type Config, type ConfigFile, type Region } from "./schema.ts";
+import { parseConfigFile, REGIONS, type Config, type ConfigFile } from "./schema.ts";
 import { ensureConfigDir, getConfigPath } from "./paths.ts";
 import { detectOutputFormat, type OutputFormat } from "../output/formatter.ts";
 import { BailianError } from "../errors/base.ts";
@@ -36,16 +36,7 @@ export function loadConfig(flags: GlobalFlags): Config {
  const accessTokenEnv = process.env.DASHSCOPE_ACCESS_TOKEN?.trim() || undefined;
  const fileAccessToken = file.access_token?.trim() || undefined;

-  const explicitRegion = (flags.region as string) || process.env.DASHSCOPE_REGION || undefined;
-  const cachedRegion = file.region;
-  const region = (explicitRegion || cachedRegion || "cn") as Region;
-
-  const baseUrl =
-    flags.baseUrl ||
-    process.env.DASHSCOPE_BASE_URL ||
-    file.base_url ||
-    REGIONS[region] ||
-    REGIONS.cn;
+  const baseUrl = flags.baseUrl || file.base_url || process.env.DASHSCOPE_BASE_URL || REGIONS.cn;

  const output: OutputFormat = detectOutputFormat(
    flags.output || process.env.DASHSCOPE_OUTPUT || file.output,
@@ -68,9 +59,7 @@ export function loadConfig(flags: GlobalFlags): Config {
    accessTokenEnv,
    fileAccessToken,
    fileApiKey,
-    fileRegion: file.region,
    configPath: getConfigPath(),
-    region,
    baseUrl,
    output,
    outputDir: file.output_dir || undefined,
@@ -84,10 +73,10 @@ export function loadConfig(flags: GlobalFlags): Config {
    accessKeySecret:
      process.env.ALIBABA_CLOUD_ACCESS_KEY_SECRET || file.access_key_secret || undefined,
    workspaceId: process.env.BAILIAN_WORKSPACE_ID || file.workspace_id || undefined,
-    consoleGatewayUrl:
-      process.env.BAILIAN_CONSOLE_GATEWAY_URL ||
-      file.console_gateway_url ||
-      "https://pre-bailian-cs.console.aliyun.com",
+    consoleSite: (flags.consoleSite as Config["consoleSite"]) || file.console_site || undefined,
+    consoleRegion: (flags.consoleRegion as string) || file.console_region || undefined,
+    consoleSwitchAgent:
+      (flags.consoleSwitchAgent as number) || file.console_switch_agent || undefined,
    verbose: flags.verbose || process.env.DASHSCOPE_VERBOSE === "1",
    quiet: flags.quiet || false,
    noColor: flags.noColor || process.env.NO_COLOR !== undefined || !process.stdout.isTTY,
@@ -18,7 +18,6 @@ export interface ConfigFile {
  api_key?: string;
  /** OAuth-style token from `bl auth login --console` callback; sent as `Authorization: Bearer …` */
  access_token?: string;
-  region?: Region;
  base_url?: string;
  output?: "text" | "json";
  output_dir?: string;
@@ -31,16 +30,18 @@ export interface ConfigFile {
  access_key_id?: string;
  access_key_secret?: string;
  workspace_id?: string;
-  console_gateway_url?: string;
+  console_site?: "domestic" | "international";
+  console_region?: string;
+  console_switch_agent?: number;
  telemetry?: boolean;
 }

-const VALID_REGIONS = new Set<string>(["cn", "us", "intl"]);
 const VALID_OUTPUTS = new Set<string>(["text", "json"]);
+const VALID_CONSOLE_SITES = new Set<string>(["domestic", "international"]);

 /**
- * A syntactically valid absolute http(s) URL. Used to validate `base_url` and
- * `console_gateway_url` from the config file: the credential-bearing client
+ * A syntactically valid absolute http(s) URL. Used to validate `base_url`
+ * from the config file: the credential-bearing client
 * sends the Bearer token to these origins, so a bare `startsWith("http")` check
 * (which also accepts e.g. "httpfoo://…") is too loose.
 */
@@ -63,8 +64,6 @@ export function parseConfigFile(raw: unknown): ConfigFile {
    out.access_token = obj.access_token;
  else if (typeof obj.accessToken === "string" && obj.accessToken.length > 0)
    out.access_token = obj.accessToken;
-  if (typeof obj.region === "string" && VALID_REGIONS.has(obj.region))
-    out.region = obj.region as Region;
  if (typeof obj.base_url === "string" && isHttpUrl(obj.base_url)) out.base_url = obj.base_url;
  if (typeof obj.output === "string" && VALID_OUTPUTS.has(obj.output))
    out.output = obj.output as ConfigFile["output"];
@@ -87,8 +86,12 @@ export function parseConfigFile(raw: unknown): ConfigFile {
    out.access_key_secret = obj.access_key_secret;
  if (typeof obj.workspace_id === "string" && obj.workspace_id.length > 0)
    out.workspace_id = obj.workspace_id;
-  if (typeof obj.console_gateway_url === "string" && isHttpUrl(obj.console_gateway_url))
-    out.console_gateway_url = obj.console_gateway_url;
+  if (typeof obj.console_site === "string" && VALID_CONSOLE_SITES.has(obj.console_site))
+    out.console_site = obj.console_site as ConfigFile["console_site"];
+  if (typeof obj.console_region === "string" && obj.console_region.length > 0)
+    out.console_region = obj.console_region;
+  if (typeof obj.console_switch_agent === "number" && obj.console_switch_agent > 0)
+    out.console_switch_agent = obj.console_switch_agent;
  if (typeof obj.telemetry === "boolean") out.telemetry = obj.telemetry;

  return out;
@@ -103,9 +106,7 @@ export interface Config {
  /** `access_token` in config file (console login). */
  fileAccessToken?: string;
  fileApiKey?: string;
-  fileRegion?: Region;
  configPath?: string;
-  region: Region;
  baseUrl: string;
  output: "text" | "json";
  outputDir?: string;
@@ -118,7 +119,9 @@ export interface Config {
  accessKeyId?: string;
  accessKeySecret?: string;
  workspaceId?: string;
-  consoleGatewayUrl: string;
+  consoleSite?: "domestic" | "international";
+  consoleRegion?: string;
+  consoleSwitchAgent?: number;
  verbose: boolean;
  quiet: boolean;
  noColor: boolean;
@@ -2,18 +2,70 @@ import type { Config } from "../config/schema.ts";
 import { BailianError } from "../errors/base.ts";
 import { ExitCode } from "../errors/codes.ts";

-const GATEWAY_ACTION = "BroadScopeAspnGateway";
 const GATEWAY_PRODUCT = "sfm_bailian";

+export type ConsoleSite = "domestic" | "international";
+
+interface ConsoleGatewayInfo {
+  csGateway: string;
+  action: string;
+}
+
+const REGION_GATEWAYS: Record<string, Record<ConsoleSite, ConsoleGatewayInfo>> = {
+  "cn-beijing": {
+    domestic: { csGateway: "bailian-cs.console.aliyun.com", action: "BroadScopeAspnGateway" },
+    international: {
+      csGateway: "bailian-cs.console.alibabacloud.com",
+      action: "BroadScopeAspnGateway",
+    },
+  },
+  "ap-southeast-1": {
+    domestic: {
+      csGateway: "modelstudio-cs.console.aliyun.com",
+      action: "IntlBroadScopeAspnGateway",
+    },
+    international: {
+      csGateway: "bailian-singapore-cs.alibabacloud.com",
+      action: "IntlBroadScopeAspnGateway",
+    },
+  },
+};
+
+function resolveGateway(region: string, site: ConsoleSite): ConsoleGatewayInfo {
+  return REGION_GATEWAYS[region]?.[site] ?? REGION_GATEWAYS["cn-beijing"]![site];
+}
+
+/** Resolved console gateway settings (same defaults as {@link callConsoleGateway}). */
+export function effectiveConsoleGatewayConfig(config: Config): {
+  consoleRegion: string;
+  consoleSite: ConsoleSite;
+  consoleSwitchAgent?: number;
+} {
+  const consoleRegion = config.consoleRegion ?? "cn-beijing";
+  const consoleSite = config.consoleSite ?? "domestic";
+  const consoleSwitchAgent = config.consoleSwitchAgent;
+  return consoleSwitchAgent != null
+    ? { consoleRegion, consoleSite, consoleSwitchAgent }
+    : { consoleRegion, consoleSite };
+}
+
 export interface ConsoleGatewayRequest {
  /** Console API name, e.g. zeldaEasy.broadscope-bailian.freeTrial.queryFreeTierQuota */
  api: string;
  data: Record<string, unknown>;
-  /** Console region (default: cn-beijing), distinct from DashScope `config.region`. */
+  /** Console region (e.g. cn-beijing, ap-southeast-1). Falls back to config.consoleRegion, then "cn-beijing". */
  region?: string;
+  /** Console site. Falls back to config.consoleSite, then "domestic". */
+  site?: ConsoleSite;
+  /** Switch-agent UID for delegated access. Falls back to config.consoleSwitchAgent. */
+  switchAgent?: number;
 }

-function buildGatewayParams(api: string, data: Record<string, unknown>): string {
+function buildGatewayParams(
+  api: string,
+  data: Record<string, unknown>,
+  switchAgent?: number,
+): string {
  return JSON.stringify({
    Api: api,
    V: "1.0",
@@ -24,6 +76,7 @@ function buildGatewayParams(api: string, data: Record<string, unknown>): string
        console: "ONE_CONSOLE",
        productCode: "p_efm",
        consoleSite: "BAILIAN_ALIYUN",
+        ...(switchAgent != null ? { switchAgent } : {}),
        ...(typeof data.cornerstoneParam === "object" && data.cornerstoneParam !== null
          ? (data.cornerstoneParam as Record<string, unknown>)
          : {}),
@@ -37,17 +90,28 @@ function buildGatewayParams(api: string, data: Record<string, unknown>): string
 * `token` is the console `access_token` (from `bl auth login --console`); when
 * omitted the request is sent without an Authorization header, which works for
 * public console APIs that don't require a login session.
+ *
+ * Gateway URL and action are resolved from `region + site` via {@link REGION_GATEWAYS}.
+ * Each parameter falls back to the corresponding config value, then to a hardcoded default.
 */
 export async function callConsoleGateway(
  config: Config,
  token: string | undefined,
-  { api, data, region = "cn-beijing" }: ConsoleGatewayRequest,
+  { api, data }: ConsoleGatewayRequest,
 ): Promise<unknown> {
-  const params = buildGatewayParams(api, data);
-  const body = new URLSearchParams({ params, region });
-  const timeoutMs = config.timeout * 1000;
+  const {
+    consoleRegion: effectiveRegion,
+    consoleSite: effectiveSite,
+    consoleSwitchAgent: effectiveSwitchAgent,
+  } = effectiveConsoleGatewayConfig(config);

-  const gatewayBase = config.consoleGatewayUrl;
+  const resolved = resolveGateway(effectiveRegion, effectiveSite);
+  const gatewayBase = `https://${resolved.csGateway}`;
+  const action = resolved.action;
+
+  const params = buildGatewayParams(api, data, effectiveSwitchAgent);
+  const body = new URLSearchParams({ params, region: effectiveRegion });
+  const timeoutMs = config.timeout * 1000;

  const headers: Record<string, string> = {
    Accept: "*/*",
@@ -56,7 +120,7 @@ export async function callConsoleGateway(
  if (token) headers.Authorization = `Bearer ${token}`;

  const res = await fetch(
-    `${gatewayBase}/cli/api.json?action=${GATEWAY_ACTION}&product=${GATEWAY_PRODUCT}&api=${encodeURIComponent(api)}`,
+    `${gatewayBase}/cli/api.json?action=${action}&product=${GATEWAY_PRODUCT}&api=${encodeURIComponent(api)}`,
    {
      method: "POST",
      headers,
@@ -74,5 +138,25 @@ export async function callConsoleGateway(
    );
  }

-  return res.json() as Promise<unknown>;
+  const json = (await res.json()) as Record<string, unknown>;
+
+  const innerData = json.data as Record<string, unknown> | undefined;
+  if (innerData?.success === false && innerData.errorCode) {
+    const errorCode = String(innerData.errorCode);
+    const notLogined = errorCode.includes("NotLogined");
+    const errorMsg = typeof innerData.errorMsg === "string" ? innerData.errorMsg : undefined;
+    throw new BailianError(
+      notLogined
+        ? "Console session is not logged in or has expired."
+        : `Console gateway error: ${errorCode}`,
+      notLogined ? ExitCode.AUTH : ExitCode.GENERAL,
+      notLogined
+        ? "Run `bl auth login --console` to sign in or refresh your console session."
+        : errorMsg && errorMsg !== errorCode
+          ? errorMsg
+          : undefined,
+    );
+  }
+
+  return json;
 }
--- a/Show More
+++ b/Show More