Files
antvis__chart-visualization…/.env.example
T
Joel Alan 833301aa2c feat: add g2 skills with best engineering practices (#36)
* chore: 新增 g2 skills

* chore: add performance part

* fix: test

* fix: Ai CR

* chore: user invocable

* chore: AI SDK 优化

* chore:  add harness

* fix: path error

* chore: 补充参考文档

* chore: 新增 grep 兜底检索

* chore: eval 自检更新

* chore: 删除无效scripts

* chore: skill 优化

* chore: add worktree

* chore: add kimi model

* chore: override env

* fix: 修复 data 格式错误

* chore: 目录结构调整

* chore: 补充评测数据

* chore: 删除 common 模块

* chore: 优化 scale skill

* chore: skill tool 移动到 cli 下

* chore: add playground

* refactor: skill cli (#37)

Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>
Co-authored-by: Joel Alan <31396322+lxfu1@users.noreply.github.com>

* test: fix ci workflow (#38)

Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>

* docs: add readme for cli usage (#39)

* docs: add readme for cli usage

* docs: fix skill typo

* docs: update

* test: fix ci

---------

Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>

* refactor: playground for cli and skill (#40)

* chore: 补全 playground libs

* fix: ts

* fix: resolve conversation

* chore: 代码优化

* fix: 修复由于 nextjs 加载源码导致 __dirname 异常 (#44)

* chore: 添加 claude plugin (#43)

* refactor: eval logic (#42)

* chore: 优化 harness 逻辑 (#48)

* chore: 优化 harness 逻辑

* chore: 优化 harness 流程

* fix: result count

* feat: 新增 retrieveWithContent (#49)

* feat: 新增 retrieveWithContent

* fix: test

* chore: loadSkillContent 目录约束

* fix: 修复 work tree 模式下 __dirname 不一致导致的优化异常

* chore: 新增 harness 优化 skill 后的基线对比逻辑 (#51)

* chore: cli 重命名 (#52)

* chore: cli 重命名

* fix: tests

* refactor: antv cli (#53)

* 精简索引构建输出并优化检索链路(移除 indexDir、使用 gray-matter 解析元数据、简化 includeContent,并统一 index 到 src/index→dist/index) (#56)

* Refactor playground AI flow to Vercel AI SDK with Skill/CLI mode tool assembly and token usage visibility (#54)

* Initial plan

* refactor(playground): migrate chat flow to vercel ai sdk

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/60eb740b-946b-4dbc-8065-b2143da12fa4

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): harden skill path handling and restore temperature

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/60eb740b-946b-4dbc-8065-b2143da12fa4

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* refactor(playground): split skill tools into separate modules

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/bcdb3e77-18a5-4b73-8c9d-4ae97f87510a

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* feat(playground): redesign dual-mode ai tools workflow

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/dd7b7fe2-54a4-4e8c-ac4d-531df443ff31

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): normalize library naming in prompts

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/dd7b7fe2-54a4-4e8c-ac4d-531df443ff31

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): tighten tool input limits and constants

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/dd7b7fe2-54a4-4e8c-ac4d-531df443ff31

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): simplify read-file validation fallback naming

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/dd7b7fe2-54a4-4e8c-ac4d-531df443ff31

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* refactor(playground): use explicit step constants and schemas

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/dd7b7fe2-54a4-4e8c-ac4d-531df443ff31

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* refactor(playground): rename modes to skill/cli and unify tools directory

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/69d26ddf-26a6-43dd-bdca-32bb851ea746

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): harden tools path handling and naming

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/69d26ddf-26a6-43dd-bdca-32bb851ea746

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): tighten tool path validation guards

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/69d26ddf-26a6-43dd-bdca-32bb851ea746

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* refactor(playground): move mode entry files above tools directory

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/e4438102-2cd2-42ac-b772-a29fa11c5a20

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* fix(playground): adapt retrieve tool to updated retrieve api

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/8d505619-d6d9-48dd-85a5-af7c42bc73b7

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* refactor(playground): remove retrieve content fallback path

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/c8187137-e11b-480a-9abf-ed20f30bb6f7

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* refactor(playground): use gray-matter in list references tool

Agent-Logs-Url: https://github.com/antvis/chart-visualization-skills/sessions/23696116-a1cb-402e-954e-06b426ddc4b2

Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>

* chore: opt cli command

* refactor: playground

* feat: fix style

* feat: update website meta info

* fix: tool call error

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>
Co-authored-by: hustcc <i@hust.cc>
Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>

* refactor: 重构 g2-chart skill (#57)

* refactor: 重构 g2-chart skill

* chore: resolve skill

* chore: 去掉 eval 中 list refercnces 逻辑

* chore: playground 模块支持加载外层 env (#58)

* chore: playground 模块支持加载外层 env

* Apply suggestions from code review

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

---------

Co-authored-by: hustcc <i@hust.cc>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* docs: add agent.md (#60)

* docs: add agent.md

* chore: fix typo

---------

Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>

* chore: 修改 env 示例 (#59)

* chore: 修改 env 示例

* chore: 新增 VL Model 配置示例

* refactor: playground (#61)

* refactor: playground

* Update playground/src/libs/cli-mode.ts

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* Update playground/src/libs/skill-mode.ts

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

* Update playground/src/app/api/generate/route.ts

Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

---------

Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

---------

Co-authored-by: hustcc <i@hust.cc>
Co-authored-by: 逍为 <xiaowei.wzw@antgroup.com>
Co-authored-by: Copilot <198982749+Copilot@users.noreply.github.com>
Co-authored-by: hustcc <7856674+hustcc@users.noreply.github.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
2026-04-14 17:04:36 +08:00

14 lines
376 B
Bash
Raw Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# AI Provider API Keys
# Playground Provider Path: playground/src/libs/provider.ts
# DeepSeek API Key
DEEPSEEK_API_KEY=your-deepseek-api-key-here
# DeepSeek API Endpoint
DEEPSEEK_API_ENDPOINT=https://api.deepseek.com
# DeepSeek API Path
DEEPSEEK_API_PATH=/v1/chat/completions
# DeepSeek Model Name
DEEPSEEK_MODEL=model-name
# VL Model NameOptional
VL_MODEL=model-name