mirror of
https://github.com/Imbad0202/academic-research-skills.git
synced 2026-09-14 13:51:17 +08:00
feat(v3.6.5): release sweep — literature_corpus[] consumer integration (PR-B) (#42)
* feat(v3.6.5): wire literature_strategist_agent corpus consumer - academic-paper/agents/literature_strategist_agent.md: add Step 0 + corpus-first flow + 4 Iron Rules + PRE-SCREENED template + F3/F4 reporting + backpointer; corpus-aware additions to Literature Matrix and Research Gap Identification. - academic-pipeline/references/literature_corpus_consumers.md: promote strategist stub block to full content; remove LINT_STUB marker and Status: Stub line; align bibliography_agent block to release wording. - scripts/corpus_consumer_manifest.json: append literature_strategist entry; update $schema_note to v3.6.5 release state. - shared/handoff_schemas.md: retire the v3.6.4 'Consumer-side integration deferred to v3.6.5+' caveat; replace with backpointer to literature_corpus_consumers.md. Lint: - check_corpus_consumer_protocol.py exit 0 (PR-B closed-set tuple match) - check_spec_consistency.py exit 0 - pytest scripts/ 331 passed Per design doc 2026-04-26-ars-v3.6.5-consumer-integration-design.md §6.2. Version sweep + CHANGELOG follows in next atomic commit per §6.3. * chore(v3.6.5): version sweep + CHANGELOG entry Atomic 7-touchpoint version bump per design doc §6.3: 1. academic-pipeline/SKILL.md: 3.6.4 → 3.6.5 (suite version invariant) 2. deep-research/SKILL.md: 2.9.1 → 2.9.2 (also syncs Version Info footer that lagged at 2.9.0 from v3.5.1 PR #36) 3. academic-paper/SKILL.md: 3.1.0 → 3.1.1 4. MODE_REGISTRY.md: v3.6.4 (2026-04-25) → v3.6.5 (2026-04-27) 5. .claude/CLAUDE.md: Skills Overview row versions, new "v3.6.5 Key Additions" section, Suite version footer; v3.6.4 deferred-caveat line rewritten as historical note. 6. README.md + README.zh-TW.md: badge, release tag URL, v3.6.5 changelog entry; v3.6.4 "Not yet shipped" caveat rewritten as historical note. 7. scripts/check_spec_consistency.py: append v3.6.5 expected pin and bump three active v3.6.4 pins (MODE_REGISTRY last-updated, CLAUDE suite version, README badge/tag); v3.3.2 – v3.6.4 regression pins retained. CHANGELOG: [3.6.5] entry per design doc §6.5 template. Verification: - check_spec_consistency.py exit 0 - check_corpus_consumer_protocol.py exit 0 - pytest scripts/ 331 passed Single atomic commit per design doc §6.3 critical note (splitting the sweep across SKILL.md, CHANGELOG, and lint causes intermediate CI red). * docs(v3.6.5): retire deferred-consumer wording in PERFORMANCE docs Codex /codex review round-1 [P3] finding: README links docs/PERFORMANCE.{md,zh-TW.md} as current guidance, but they still claimed "no ARS agent reads literature_corpus[]" and "consumer-side integration deferred to v3.6.5+" — directly contradicting the v3.6.5 release state shipped in the rest of this branch. - docs/PERFORMANCE.md §"Consumer-side integration": rewritten to describe v3.6.5 bibliography_agent + literature_strategist_agent corpus-first flow, four Iron Rules, PRE-SCREENED block, presence-based auto-engagement, parse-failure fallback, and citation_compliance_agent v3.6.6+ deferral; backpointer to literature_corpus_consumers.md. - docs/PERFORMANCE.zh-TW.md §"消費端整合": parallel rewrite. - Both files: "What v3.6.4 does NOT do" / "v3.6.4 不做的事" subheading renamed to "Ingestion-layer boundaries" / "導入層邊界" — the listed boundaries (no PDF ingest, no live API, no paywalled content) are still true at v3.6.5 for the ingestion layer; the version-bound framing was misleading. Pattern from feedback_codex_review_per_schema_increment.md and feedback_codex_preflight_for_writing.md: codex catches cross-doc staleness that both Claude and lint miss because docs/PERFORMANCE wasn't in the design doc §6.3 seven-touchpoint sweep but is reachable from README as current guidance. Verification: - check_spec_consistency.py exit 0 (markdown link checker re-runs on PERFORMANCE) - check_corpus_consumer_protocol.py exit 0 - pytest scripts/ 331 passed * docs(v3.6.5): promote consumer protocol header from pre-release to release Codex /codex review round-2 [P3]: literature_corpus_consumers.md still opened with "Pre-release in PR-A (single consumer wired); full v3.6.5 release pending PR-B" while the same branch ships the full v3.6.5 release state across manifest, CHANGELOG, README, SKILL.md, and Schema 9. Readers following the release/manifest path would have hit a contradiction. Header now: "Released in v3.6.5 (both Phase 1 consumers wired)". All three lints + 331 tests still green. * chore(v3.6.5): drop order-coupled "(see above)" forward-pointer /simplify quality review finding: .claude/CLAUDE.md v3.6.4 section's bullet "...were wired in v3.6.5 (see above)" had two issues — the "(see above)" parenthetical is order-coupled (silently breaks if v3.6.4 ever moves above v3.6.5 in section ordering, and no lint catches it), and the bullet itself mostly restates what the new v3.6.5 Key Additions section already says. Aligned to the cleaner phrasing already in README.md ("...consumer integration landed in v3.6.5"). No "(see above)" anchor; tells the historical fact about v3.6.4's input-port-only shape, lets v3.6.5 section speak for itself. All three lints + 331 tests still green.
This commit is contained in:
Edward Cheng-I Wu
committed by
GitHub
GitHub
parent
88b3195680
commit
3e953ce029
+15
-6
@@ -6,10 +6,19 @@ A suite of Claude Code skills for rigorous academic research, paper writing, pee
|
||||
|
||||
| Skill | Purpose | Key Modes |
|
||||
|-------|---------|-----------|
|
||||
| `deep-research` v2.9.1 | 13-agent research team | full, quick, socratic, review, lit-review, fact-check, systematic-review |
|
||||
| `academic-paper` v3.1.0 | 12-agent paper writing | full, plan, outline-only, revision, revision-coach, abstract-only, lit-review, format-convert, citation-check, disclosure |
|
||||
| `deep-research` v2.9.2 | 13-agent research team | full, quick, socratic, review, lit-review, fact-check, systematic-review |
|
||||
| `academic-paper` v3.1.1 | 12-agent paper writing | full, plan, outline-only, revision, revision-coach, abstract-only, lit-review, format-convert, citation-check, disclosure |
|
||||
| `academic-paper-reviewer` v1.9.0 | Multi-perspective paper review (5 reviewers + optional cross-model DA critique) | full, re-review, quick, methodology-focus, guided, calibration |
|
||||
| `academic-pipeline` v3.6.4 | Full pipeline orchestrator | (coordinates all above) |
|
||||
| `academic-pipeline` v3.6.5 | Full pipeline orchestrator | (coordinates all above) |
|
||||
|
||||
## v3.6.5 Key Additions
|
||||
|
||||
- **Material Passport `literature_corpus[]` consumer integration in Phase 1**: `deep-research/agents/bibliography_agent.md` and `academic-paper/agents/literature_strategist_agent.md` now read `literature_corpus[]` via the **corpus-first, search-fills-gap** flow when the passport carries a non-empty corpus. Both consumers follow the same five-step shared flow (Step 0 presence detection → Step 1 pre-screen → Step 2 search-fills-gap → Step 3 merge → Step 4 emit Search Strategy report) and the same four Iron Rules (Same criteria / No silent skip / No corpus mutation / Graceful fallback on parse failure).
|
||||
- **PRE-SCREENED reproducibility block**: Search Strategy reports gain a PRE-SCREENED FROM USER CORPUS block enumerating included / excluded / skipped corpus entries, with F3 zero-hit note and F4a–F4f provenance reporting that compose around partial declaration of `obtained_via` / `obtained_at`. `final_included = pre_screened_included[] ∪ external_included[]` stays neutral — no provenance tags on bibliography entries or literature matrix rows.
|
||||
- **Consumer protocol reference**: `academic-pipeline/references/literature_corpus_consumers.md` carries the canonical PRE-SCREENED template, BAD/GOOD examples, four Iron Rules, and per-consumer reading instructions. Both consumer agents backpoint to this reference.
|
||||
- **CI lint** `scripts/check_corpus_consumer_protocol.py` enforcing nine protocol invariants with manifest-driven consumer list (`scripts/corpus_consumer_manifest.json`).
|
||||
- **Schema 9 caveat retired**: `shared/handoff_schemas.md` retired the v3.6.4 "Consumer-side integration deferred to v3.6.5+" caveat; replaced with backpointer to the consumer protocol.
|
||||
- **No schema change**: existing user adapters work without modification. Consumer integration is presence-based: auto-engages when passport carries a non-empty `literature_corpus[]` and parses cleanly. Parse failures fall back to external-DB-only flow with a `[CORPUS PARSE FAILURE]` surface. No new env flag introduced. `citation_compliance_agent` corpus integration deferred to v3.6.6+.
|
||||
|
||||
## v3.6.4 Key Additions
|
||||
|
||||
@@ -18,7 +27,7 @@ A suite of Claude Code skills for rigorous academic research, paper writing, pee
|
||||
- **Three reference Python adapters**: `scripts/adapters/{folder_scan,zotero,obsidian}.py` with tests and fixtures. Starting points only; users are expected to write their own adapters for non-reference corpus sources.
|
||||
- **Rejection log contract**: `shared/contracts/passport/rejection_log.schema.json`. Always emitted, empty when no rejections; closed enum of categorical reason values.
|
||||
- **CI lint + pytest job**: `scripts/check_literature_corpus_schema.py` validates schemas + examples; `scripts/sync_adapter_docs.py --check` prevents schema→docs drift; new `pytest.yml` workflow runs `scripts/adapters/tests/` on path-filtered triggers.
|
||||
- **Not yet shipped**: no ARS agent reads `literature_corpus[]` yet. Consumer-side integration deferred to v3.6.5+. v3.6.4 defines the input port only.
|
||||
- **Input-port-only at v3.6.4**: v3.6.4 shipped the schema and adapter contract; consumer integration landed in v3.6.5.
|
||||
|
||||
## v3.6.3 Key Additions
|
||||
|
||||
@@ -115,7 +124,7 @@ Materials: Complete paper text. field_analyst_agent auto-detects domain and conf
|
||||
Materials: Editorial Decision Letter, Revision Roadmap, Per-reviewer detailed comments
|
||||
|
||||
## Version Info
|
||||
- **Suite version**: 3.6.4 (per CHANGELOG.md)
|
||||
- **Last Updated**: 2026-04-25
|
||||
- **Suite version**: 3.6.5 (per CHANGELOG.md)
|
||||
- **Last Updated**: 2026-04-27
|
||||
- **Author**: Cheng-I Wu
|
||||
- **License**: CC-BY-NC 4.0
|
||||
|
||||
@@ -4,6 +4,46 @@ All notable changes to this project will be documented in this file.
|
||||
|
||||
## [Unreleased]
|
||||
|
||||
## [3.6.5] - 2026-04-27
|
||||
|
||||
### Added
|
||||
|
||||
- Material Passport `literature_corpus[]` consumer integration in Phase 1
|
||||
(deep-research/bibliography_agent + academic-paper/literature_strategist_agent).
|
||||
Corpus-first, search-fills-gap flow with PRE-SCREENED reproducibility block.
|
||||
Reproducibility for systematic-review use is preserved through Iron Rule 1
|
||||
same-criteria parity plus Step 2 case C (standard external search runs even
|
||||
when corpus fully covers RQ subtopics).
|
||||
- `academic-pipeline/references/literature_corpus_consumers.md` — consumer protocol
|
||||
reference with four Iron Rules (Same criteria / No silent skip / No corpus mutation /
|
||||
Graceful fallback on parse failure) and per-consumer reading instructions.
|
||||
- `scripts/check_corpus_consumer_protocol.py` — CI lint enforcing nine protocol invariants
|
||||
with manifest-driven consumer list and stub-block opt-out.
|
||||
- `scripts/corpus_consumer_manifest.json` — supported-consumer manifest.
|
||||
|
||||
### Changed
|
||||
|
||||
- `shared/handoff_schemas.md` Schema 9 — retired the v3.6.4 "Consumer-side integration
|
||||
deferred to v3.6.5+" caveat; replaced with backpointer to the consumer protocol.
|
||||
- `deep-research/SKILL.md` 2.9.1 → 2.9.2 — bibliography_agent corpus-first flow (also
|
||||
syncs Version Info footer that lagged at 2.9.0).
|
||||
- `academic-paper/SKILL.md` 3.1.0 → 3.1.1 — literature_strategist_agent corpus-first flow.
|
||||
- `academic-pipeline/SKILL.md` 3.6.4 → 3.6.5 — suite version invariant.
|
||||
- `.claude/CLAUDE.md`, `MODE_REGISTRY.md`, `README.md`, `README.zh-TW.md`,
|
||||
`scripts/check_spec_consistency.py` updated for the version bump (suite version,
|
||||
badge, tag, changelog heading).
|
||||
|
||||
### Notes
|
||||
|
||||
- Consumer integration is presence-based: auto-engages when passport carries a
|
||||
non-empty `literature_corpus[]` and parses cleanly. Parse failures fall back
|
||||
to external-DB-only flow with a `[CORPUS PARSE FAILURE]` surface. No new env
|
||||
flag introduced.
|
||||
- Schema is unchanged from v3.6.4. Existing user adapters work without modification.
|
||||
- `citation_compliance_agent` corpus integration deferred to v3.6.6+.
|
||||
- `source_pointer` is not dereferenced by consumers; URI resolution remains a future
|
||||
`source_verification_agent` concern.
|
||||
|
||||
## [3.6.4] - 2026-04-25
|
||||
|
||||
### Added
|
||||
|
||||
+1
-1
@@ -4,7 +4,7 @@ Single source of truth for all modes across the ARS suite. **25 modes** across 4
|
||||
|
||||
When adding or modifying modes, update this file first — SKILL.md files and CLAUDE.md should reference this registry.
|
||||
|
||||
Last updated: v3.6.4 (2026-04-25)
|
||||
Last updated: v3.6.5 (2026-04-27)
|
||||
|
||||
---
|
||||
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Academic Research Skills for Claude Code
|
||||
|
||||
[](https://github.com/Imbad0202/academic-research-skills/releases/tag/v3.6.4)
|
||||
[](https://github.com/Imbad0202/academic-research-skills/releases/tag/v3.6.5)
|
||||
[](https://creativecommons.org/licenses/by-nc/4.0/)
|
||||
[](https://buymeacoffee.com/crucify020v)
|
||||
|
||||
@@ -289,6 +289,16 @@ https://github.com/Imbad0202/academic-research-skills
|
||||
|
||||
## Changelog
|
||||
|
||||
### v3.6.5 (2026-04-27) — Material Passport `literature_corpus[]` Consumer Integration
|
||||
|
||||
- **Two Phase 1 literature consumers** wired: `deep-research/agents/bibliography_agent.md` and `academic-paper/agents/literature_strategist_agent.md`. Both follow the same five-step **corpus-first, search-fills-gap** flow when the passport carries a non-empty `literature_corpus[]` and the same four Iron Rules (Same criteria / No silent skip / No corpus mutation / Graceful fallback on parse failure).
|
||||
- **PRE-SCREENED reproducibility block** in Search Strategy reports: enumerates included / excluded / skipped corpus entries, with F3 zero-hit note and F4a–F4f provenance reporting that compose around partial declaration of `obtained_via` / `obtained_at`. `final_included = pre_screened_included[] ∪ external_included[]` stays neutral — no provenance tags on bibliography entries or literature matrix rows.
|
||||
- **Consumer protocol reference** at `academic-pipeline/references/literature_corpus_consumers.md` with the canonical PRE-SCREENED template, BAD/GOOD examples, four Iron Rules, and per-consumer reading instructions.
|
||||
- **CI lint** `scripts/check_corpus_consumer_protocol.py` enforcing nine protocol invariants with manifest-driven consumer list (`scripts/corpus_consumer_manifest.json`).
|
||||
- **Schema 9 caveat retired**: `shared/handoff_schemas.md` retired the v3.6.4 "Consumer-side integration deferred to v3.6.5+" caveat; replaced with backpointer to the consumer protocol.
|
||||
- Presence-based, no schema change, no new env flag. Parse failures fall back to external-DB-only flow with a `[CORPUS PARSE FAILURE]` surface. `citation_compliance_agent` corpus integration deferred to v3.6.6+.
|
||||
- No breaking changes. Existing user adapters work without modification.
|
||||
|
||||
### v3.6.4 (2026-04-25) — Material Passport `literature_corpus[]` Input Port
|
||||
|
||||
- **`literature_corpus[]` field** added to Schema 9 as an optional input port for user-owned literature. Each entry conforms to `shared/contracts/passport/literature_corpus_entry.schema.json` (CSL-JSON authors, year, title, source_pointer + private optional `abstract` / `user_notes`).
|
||||
@@ -296,7 +306,7 @@ https://github.com/Imbad0202/academic-research-skills
|
||||
- **Three reference Python adapters** under `scripts/adapters/`: `folder_scan.py` (filesystem of PDFs), `zotero.py` (Better BibTeX JSON export), `obsidian.py` (vault frontmatter). Starting points only; users are expected to write their own adapters for non-reference sources.
|
||||
- **Rejection log contract** at `shared/contracts/passport/rejection_log.schema.json` with closed enum of categorical reason values; always emitted (empty when no rejections).
|
||||
- **CI gates**: `scripts/check_literature_corpus_schema.py` validates schemas + adapter examples; `scripts/sync_adapter_docs.py --check` prevents schema→docs drift; new `pytest.yml` workflow runs `scripts/adapters/tests/` on path-filtered triggers.
|
||||
- **Not yet shipped**: no ARS agent reads `literature_corpus[]` yet. Consumer-side integration is deferred to v3.6.5+. v3.6.4 defines the input port only.
|
||||
- **Input-port-only at v3.6.4**: v3.6.4 shipped the schema and adapter contract without consumer integration. `bibliography_agent` and `literature_strategist_agent` were wired in v3.6.5.
|
||||
- No breaking changes.
|
||||
|
||||
### v3.6.3 (2026-04-23) — Opt-in Passport Reset Boundary
|
||||
|
||||
+12
-2
@@ -1,6 +1,6 @@
|
||||
# Academic Research Skills for Claude Code
|
||||
|
||||
[](https://github.com/Imbad0202/academic-research-skills/releases/tag/v3.6.4)
|
||||
[](https://github.com/Imbad0202/academic-research-skills/releases/tag/v3.6.5)
|
||||
[](https://creativecommons.org/licenses/by-nc/4.0/)
|
||||
[](https://buymeacoffee.com/crucify020v)
|
||||
|
||||
@@ -270,6 +270,16 @@ https://github.com/Imbad0202/academic-research-skills
|
||||
|
||||
## 更新紀錄
|
||||
|
||||
### v3.6.5（2026-04-27）— Material Passport `literature_corpus[]` Consumer 整合
|
||||
|
||||
- **Phase 1 兩個文獻 consumer** 接上：`deep-research/agents/bibliography_agent.md` 與 `academic-paper/agents/literature_strategist_agent.md`。當 passport 帶有非空 `literature_corpus[]` 時，兩者都走相同的五步 **corpus-first、search-fills-gap** 流程，並遵守相同的四條 Iron Rule（Same criteria / No silent skip / No corpus mutation / Graceful fallback on parse failure）。
|
||||
- **PRE-SCREENED 可重現區塊** 進 Search Strategy 報告：列出已納入／排除／略過的 corpus entry，附 F3 zero-hit 註解與 F4a–F4f provenance 報告（針對 `obtained_via` / `obtained_at` 部分宣告情境）。`final_included = pre_screened_included[] ∪ external_included[]` 維持 neutral — bibliography entry 與 literature matrix row 不掛 provenance 標籤。
|
||||
- **Consumer 協定參考文件** 在 `academic-pipeline/references/literature_corpus_consumers.md`，包含 PRE-SCREENED 模板、BAD/GOOD 範例、四條 Iron Rule 與 per-consumer 讀取指示。
|
||||
- **CI lint** `scripts/check_corpus_consumer_protocol.py` 透過 manifest 驅動的 consumer 清單（`scripts/corpus_consumer_manifest.json`）強制九條協定不變式。
|
||||
- **Schema 9 caveat 退役**：`shared/handoff_schemas.md` 移除 v3.6.4「Consumer-side integration deferred to v3.6.5+」一行，改成指向 consumer 協定的 backpointer。
|
||||
- 採 presence-based 啟動，不變更 schema、不引入新 env flag。Parse 失敗 fallback 到 external-DB-only flow，並 surface `[CORPUS PARSE FAILURE]`。`citation_compliance_agent` 的 corpus 整合延到 v3.6.6+。
|
||||
- 無破壞性變更，既有使用者 adapter 不需修改。
|
||||
|
||||
### v3.6.4（2026-04-25）— Material Passport `literature_corpus[]` 輸入埠
|
||||
|
||||
- **Schema 9 新增 `literature_corpus[]`** 選填欄位作為使用者文獻的輸入埠。每筆 entry 符合 `shared/contracts/passport/literature_corpus_entry.schema.json`（CSL-JSON authors / year / title / source_pointer，加上 PRIVATE 選填的 `abstract` / `user_notes`）。
|
||||
@@ -277,7 +287,7 @@ https://github.com/Imbad0202/academic-research-skills
|
||||
- **三個 reference Python adapter** 在 `scripts/adapters/`：`folder_scan.py`（檔案系統的 PDF 資料夾）、`zotero.py`（Better BibTeX JSON export）、`obsidian.py`（vault frontmatter）。僅供起點參考；非 reference source 預期使用者自行實作 adapter。
|
||||
- **Rejection log 契約** 在 `shared/contracts/passport/rejection_log.schema.json`，採用封閉 enum 的 categorical reason 值；永遠輸出（無 rejection 時為空）。
|
||||
- **CI 把關**：`scripts/check_literature_corpus_schema.py` 驗 schemas + adapter examples；`scripts/sync_adapter_docs.py --check` 防 schema→docs drift；新 `pytest.yml` workflow 在 path-filtered 觸發跑 `scripts/adapters/tests/`。
|
||||
- **尚未推進的部分**：v3.6.4 還沒有任何 ARS agent 讀 `literature_corpus[]`。Consumer-side 整合延到 v3.6.5+，v3.6.4 只定義輸入埠。
|
||||
- **僅輸入埠**：v3.6.4 只定義 schema 與 adapter 契約，consumer 整合到 v3.6.5 才接上 `bibliography_agent` 與 `literature_strategist_agent`。
|
||||
- 無破壞性變更。
|
||||
|
||||
### v3.6.3（2026-04-23）— 選用式 Passport 重置邊界
|
||||
|
||||
@@ -2,8 +2,8 @@
|
||||
name: academic-paper
|
||||
description: "12-agent academic paper writing pipeline. 10 modes (full/plan/outline/revision/revision-coach/abstract/lit-review/format-convert/citation-check/disclosure). 6 paper types, 5 citation formats, bilingual abstracts, LaTeX/DOCX-via-Pandoc/PDF output. Style Calibration + Writing Quality Check + Anti-Patterns with IRON RULE markers. Triggers: write paper, academic paper, guide my paper, parse reviews, AI disclosure, 寫論文, 學術論文, 引導我寫論文, 審查意見."
|
||||
metadata:
|
||||
version: "3.1.0"
|
||||
last_updated: "2026-04-20"
|
||||
version: "3.1.1"
|
||||
last_updated: "2026-04-27"
|
||||
status: active
|
||||
data_access_level: redacted
|
||||
task_type: open-ended
|
||||
@@ -318,8 +318,8 @@ academic-paper + academic-paper-reviewer -> Peer review -> revision loop
|
||||
|
||||
| Item | Content |
|
||||
|------|---------|
|
||||
| Skill Version | 3.1.0 |
|
||||
| Last Updated | 2026-04-20 |
|
||||
| Skill Version | 3.1.1 |
|
||||
| Last Updated | 2026-04-27 |
|
||||
| Maintainer | Cheng-I Wu |
|
||||
| Dependent Skills | deep-research v1.0+ (upstream), academic-paper-reviewer v1.0+ (downstream) |
|
||||
|
||||
|
||||
@@ -101,6 +101,8 @@ Create a Source x Theme matrix:
|
||||
| Author3 (Year) | | x | x | main | Mixed | High |
|
||||
```
|
||||
|
||||
When the Material Passport carries a non-empty `literature_corpus[]` and the corpus-first flow ran (see §"Reading `literature_corpus[]` from Material Passport"), the matrix is built over `final_included = pre_screened_included[] ∪ external_included[]`. Source rows stay neutral — no provenance column distinguishes corpus from external entries. Provenance accounting lives in the PRE-SCREENED block of the Search Strategy report, not in the matrix.
|
||||
|
||||
## Research Gap Identification
|
||||
|
||||
After reviewing the literature, identify:
|
||||
@@ -112,6 +114,8 @@ After reviewing the literature, identify:
|
||||
|
||||
-> These gaps inform the paper's contribution statement.
|
||||
|
||||
When the corpus-first flow ran, gap identification operates over the merged `final_included` set. The PRE-SCREENED block's zero-hit note (F3) and `uncovered_topics` from Step 2 case A / B' surface coverage gaps that originated in corpus screening; carry those forward into this section so user-curated coverage limits become explicit research-gap claims rather than silent omissions.
|
||||
|
||||
## Output Format
|
||||
|
||||
```markdown
|
||||
@@ -144,6 +148,132 @@ After reviewing the literature, identify:
|
||||
| Discussion | Author2, Author7 |
|
||||
```
|
||||
|
||||
## Reading `literature_corpus[]` from Material Passport (v3.6.5+)
|
||||
|
||||
**Backpointer**: see [`academic-pipeline/references/literature_corpus_consumers.md`](../../academic-pipeline/references/literature_corpus_consumers.md) for the full consumer protocol, BAD/GOOD examples, and shared template.
|
||||
|
||||
When the input Material Passport carries a non-empty `literature_corpus[]`, this agent enters the **corpus-first, search-fills-gap** flow. The flow has five steps and four Iron Rules; the PRE-SCREENED block makes corpus utilisation reproducible. The merged `final_included` set feeds the Annotated Bibliography, Literature Matrix, Research Gap Identification, and Recommended Sources by Paper Section sections above without altering their formats.
|
||||
|
||||
### The four Iron Rules
|
||||
|
||||
1. **Iron Rule 1 — Same criteria.** Apply the same Inclusion / Exclusion criteria (§"Step 4: Inclusion/Exclusion Criteria") to corpus entries and external database results. No exceptions.
|
||||
2. **Iron Rule 2 — No silent skip.** Any skipped corpus entry must be recorded in the PRE-SCREENED block's skipped sub-section with a reason. Silently dropping an entry is a prompt-layer violation.
|
||||
3. **Iron Rule 3 — No corpus mutation.** Consumer agents never modify, backfill, or derive new content into `literature_corpus[]`. Read only.
|
||||
4. **Iron Rule 4 — Graceful fallback on parse failure.** Consumer agents do NOT re-validate schema, do NOT parse JSON Schema at runtime, and do NOT dereference `source_pointer` URIs. When the corpus cannot be parsed, emit `[CORPUS PARSE FAILURE: <cause>]` and fall back to external-DB-only flow.
|
||||
|
||||
### Step 0: presence detection and minimal shape
|
||||
|
||||
The agent applies a MINIMAL SHAPE CHECK on the corpus before reading further. This is not JSON Schema validation. It checks only what the consumer needs to read each entry safely — the v3.6.4 required fields:
|
||||
|
||||
- shape OK ≡ `literature_corpus` is a YAML list AND
|
||||
- each entry is a YAML mapping AND
|
||||
- each entry has `citation_key` (non-empty string), `title` (non-empty string), `authors` (non-empty list), `year` (numeric-coercible), `source_pointer` (non-empty string).
|
||||
|
||||
If the passport lacks `literature_corpus` or it is empty, run the original 4-Layer Progressive Strategy (§"Detailed Execution Algorithm") unchanged. If parse or shape check fails, emit `[CORPUS PARSE FAILURE: <one-line cause>]` and fall back. Otherwise, continue to Step 1.
|
||||
|
||||
### Step 1: pre-screen corpus against current RQ
|
||||
|
||||
For each entry:
|
||||
|
||||
1. Read the five required fields and any optional fields present (`venue`, `doi`, `tags`, `abstract`, `user_notes`).
|
||||
2. Apply the current Inclusion / Exclusion criteria (peer-review status, date range, language, relevance) to whatever fields are present. `title` is always available; `abstract` and `tags` participate only when populated. Field absence narrows the screening surface but never causes SKIP.
|
||||
3. Classify as INCLUDE / EXCLUDE / SKIP. SKIP fires only when criteria cannot be applied at all (see F1 in spec §4.1).
|
||||
|
||||
The Phase A title/abstract screening described in §"Source Screening Protocol" applies to corpus entries identically; the difference is only that the input list is the user's curated corpus rather than the Layer 1-4 hit set.
|
||||
|
||||
### Step 2: search-fills-gap (external DB)
|
||||
|
||||
```
|
||||
derive uncovered_topics = RQ subtopics − {topics covered by pre_screened_included[]}
|
||||
user_corpus_only = user explicitly asked "use my corpus only"
|
||||
|
||||
case A: uncovered_topics non-empty AND NOT user_corpus_only
|
||||
→ external DB search scoped to uncovered_topics, run via 4-Layer Progressive Strategy
|
||||
case B: uncovered_topics empty AND user_corpus_only
|
||||
→ skip external; surface "external search omitted on user request"
|
||||
case B': uncovered_topics non-empty AND user_corpus_only
|
||||
→ skip external BUT surface uncovered_topics as known coverage gap
|
||||
case C: uncovered_topics empty AND NOT user_corpus_only
|
||||
→ standard external search (not scope-limited; newer-work + dedup validation)
|
||||
```
|
||||
|
||||
The external search executes the 4-Layer Progressive Strategy (Boolean → Citation Chaining → Forward Tracking → Semantic). Iron Rule 1 applies — the same Inclusion/Exclusion criteria screen Layer 1-4 hits as screened corpus entries.
|
||||
|
||||
### Step 3: merge
|
||||
|
||||
`final_included = pre_screened_included[] ∪ external_included[]`. The annotated bibliography stays neutral — no source-attribution tags on entries, no provenance column in the Literature Matrix.
|
||||
|
||||
### Step 4: emit Search Strategy Report
|
||||
|
||||
The PRE-SCREENED block goes into the Search Strategy section of the Output Format above, immediately before the existing `Databases` line.
|
||||
|
||||
### PRE-SCREENED block template
|
||||
|
||||
```markdown
|
||||
PRE-SCREENED FROM USER CORPUS:
|
||||
- Adapter: <obtained_via enum value | "<unspecified>" | "mixed (...)">
|
||||
# e.g., zotero-bbt-export, or "<unspecified>" per F4a,
|
||||
# or "<value> (N of M entries declared)" per F4b,
|
||||
# or "mixed (zotero-bbt-export: K, ..., undeclared: U)" per F4c
|
||||
- Snapshot date: <max(obtained_at)> # ISO 8601, or "<unspecified>" per F4d,
|
||||
# or "<date> (M of N entries declared)" per F4e,
|
||||
# or append "(spans <N> days; corpus may not be a single snapshot)" per F4f
|
||||
- Total entries scanned: <N>
|
||||
- Pre-screening result:
|
||||
- Included: <K> entries
|
||||
citation_keys:
|
||||
- <k1>
|
||||
- <k2>
|
||||
- Excluded by inclusion / exclusion criteria: <E> entries
|
||||
citation_keys:
|
||||
- <e1>
|
||||
(omit this sub-block if 0)
|
||||
- Skipped (criteria cannot be applied): <S> entries
|
||||
citation_keys with reasons:
|
||||
- <key>: <reason>
|
||||
(omit this sub-block if 0)
|
||||
- Zero-hit note (emit per F3 only when Included: 0):
|
||||
Zero-hit note (corpus non-empty, 0 included after screening): possible
|
||||
causes are (a) corpus is stale relative to current RQ, (b) RQ has
|
||||
shifted away from what the user originally curated, (c) adapter
|
||||
exported entries unrelated to this RQ.
|
||||
- Note: presence in corpus does not imply inclusion;
|
||||
same criteria applied to corpus and external sources.
|
||||
```
|
||||
|
||||
Lists with more than 50 entries truncate to first 20 + last 5 alphabetically, with an appendix file at `pre_screened_citation_keys_<list>_<timestamp>.txt`. Skipped truncation preserves `<key>: <reason>` in both inline and appendix forms. See spec §3.2 for the full truncation rule.
|
||||
|
||||
### Zero-hit and provenance reporting (F3 / F4)
|
||||
|
||||
Two reproducibility surfaces sit inside the PRE-SCREENED block. The agent emits each one when the corresponding trigger fires; both are non-blocking.
|
||||
|
||||
**Zero-hit note (F3).** When `pre_screened_included[]` is empty after Step 1 — corpus is non-empty but no entry survived screening — the agent emits a zero-hit note inside the PRE-SCREENED block listing the three plausible causes:
|
||||
|
||||
```
|
||||
- Zero-hit note (corpus non-empty, 0 included after screening): possible causes
|
||||
are (a) corpus is stale relative to current RQ, (b) RQ has shifted away from
|
||||
what the user originally curated, (c) adapter exported entries unrelated to
|
||||
this RQ.
|
||||
```
|
||||
|
||||
The note appears regardless of which Step 2 case fires next. Step 2 dispatch follows F3 in spec §4.1: NOT user_corpus_only routes through case A or C with external DB; user_corpus_only routes through case B' with no external search but explicit gap surfacing.
|
||||
|
||||
**Provenance reporting (F4a–F4f).** `obtained_via` and `obtained_at` are optional in v3.6.4. The PRE-SCREENED block's `Adapter:` and `Snapshot date:` lines must reflect actual coverage, not invent enum values:
|
||||
|
||||
| Sub-case | Trigger | `Adapter:` line content |
|
||||
|---|---|---|
|
||||
| F4a | Zero entries declare `obtained_via` | `Adapter: <unspecified>` + trailing note `Adapter origin not declared; user-written adapter should populate obtained_via per v3.6.4 schema recommendation.` |
|
||||
| F4b | At least one entry declares; all declared share single value | `Adapter: <enum value> (N of M entries declared)` |
|
||||
| F4c | Two or more distinct enum values among declared entries | `Adapter: mixed (zotero-bbt-export: K, obsidian-vault: L, ..., undeclared: U)` |
|
||||
|
||||
| Sub-case | Trigger | `Snapshot date:` line content |
|
||||
|---|---|---|
|
||||
| F4d | Zero entries declare `obtained_at` | `Snapshot date: <unspecified>` + trailing note `Snapshot date not declared; reproducibility is reduced. Adapter should populate obtained_at per v3.6.4 schema recommendation.` |
|
||||
| F4e | Partial coverage | `Snapshot date: <max(obtained_at)> (M of N entries declared)` |
|
||||
| F4f | Wide spread (>90 days between min and max) | append `(spans <N> days; corpus may not be a single snapshot)`. Composes with F4e. |
|
||||
|
||||
F4a/b/c are mutually exclusive by trigger. F4d applies only when zero entries declare `obtained_at`; F4e and F4f compose. Never silently fill in or guess; never demand presence. See spec §4.2 for the full precedence reasoning.
|
||||
|
||||
## Detailed Execution Algorithm
|
||||
|
||||
### Complete Search Workflow (4-Layer Progressive Strategy)
|
||||
|
||||
@@ -2,8 +2,8 @@
|
||||
name: academic-pipeline
|
||||
description: "Orchestrator for the full academic research pipeline: research -> write -> integrity check -> review -> revise -> re-review -> re-revise -> final integrity check -> finalize. Coordinates deep-research, academic-paper, and academic-paper-reviewer into a seamless 10-stage workflow with mandatory integrity verification, two-stage peer review, and reproducible quality gates. Triggers on: academic pipeline, research to paper, full paper workflow, paper pipeline, end-to-end paper, research-to-publication, complete paper workflow."
|
||||
metadata:
|
||||
version: "3.6.4"
|
||||
last_updated: "2026-04-25"
|
||||
version: "3.6.5"
|
||||
last_updated: "2026-04-27"
|
||||
depends_on: "deep-research, academic-paper, academic-paper-reviewer"
|
||||
status: active
|
||||
data_access_level: verified_only
|
||||
@@ -14,7 +14,7 @@ metadata:
|
||||
- academic-paper-reviewer
|
||||
---
|
||||
|
||||
# Academic Pipeline v3.6.4 — Full Academic Research Workflow Orchestrator
|
||||
# Academic Pipeline v3.6.5 — Full Academic Research Workflow Orchestrator
|
||||
|
||||
A lightweight orchestrator that manages the complete academic pipeline from research exploration to final manuscript. It does not perform substantive work — it only detects stages, recommends modes, dispatches skills, manages transitions, and tracks state.
|
||||
|
||||
@@ -579,8 +579,8 @@ Stage 5: academic-paper (format-convert mode)
|
||||
|
||||
| Item | Content |
|
||||
|------|---------|
|
||||
| Skill Version | 3.6.4 |
|
||||
| Last Updated | 2026-04-25 |
|
||||
| Skill Version | 3.6.5 |
|
||||
| Last Updated | 2026-04-27 |
|
||||
| Maintainer | Cheng-I Wu |
|
||||
| Dependent Skills | deep-research v2.0+, academic-paper v2.0+, academic-paper-reviewer v1.1+ |
|
||||
| Role | Full academic research workflow orchestrator |
|
||||
|
||||
@@ -1,6 +1,6 @@
|
||||
# Consumer Protocol — `literature_corpus[]` Reading
|
||||
|
||||
**Status**: Pre-release in PR-A (single consumer wired); full v3.6.5 release pending PR-B
|
||||
**Status**: Released in v3.6.5 (both Phase 1 consumers wired)
|
||||
**Applies to**: any agent in this repo that reads `literature_corpus[]` from a Material Passport
|
||||
**Authoritative spec**: [`docs/design/2026-04-26-ars-v3.6.5-consumer-integration-design.md`](../../docs/design/2026-04-26-ars-v3.6.5-consumer-integration-design.md)
|
||||
|
||||
@@ -139,7 +139,7 @@ F4a/b/c are mutually exclusive by trigger. F4d applies only when zero entries de
|
||||
|
||||
## Consumer: bibliography_agent
|
||||
|
||||
**Status**: Wired in PR-A (pre-release of v3.6.5; full release lands with PR-B)
|
||||
**Status**: Wired in v3.6.5
|
||||
**Skill**: deep-research
|
||||
**Phase**: 1 (literature search and curation)
|
||||
**Agent file**: [`deep-research/agents/bibliography_agent.md`](../../deep-research/agents/bibliography_agent.md)
|
||||
@@ -150,9 +150,11 @@ When `literature_corpus[]` is non-empty and parses cleanly, the agent enters Ste
|
||||
|
||||
## Consumer: literature_strategist_agent
|
||||
|
||||
**Status:** Stub — implementation in PR-B (v3.6.5)
|
||||
<!-- LINT_STUB: skip_cross_check -->
|
||||
**Status**: Wired in v3.6.5
|
||||
**Skill**: academic-paper
|
||||
**Phase**: 1 (literature search and curation)
|
||||
**Agent file**: [`academic-paper/agents/literature_strategist_agent.md`](../../academic-paper/agents/literature_strategist_agent.md)
|
||||
|
||||
The academic-paper literature_strategist agent will follow the same shared reading flow when PR-B ships. PR-A intentionally does not modify the strategist agent; this stub block exists so the manifest cross-check (L2 sub-invariant) has a target without flagging the strategist as a falsely-shipped consumer.
|
||||
The academic-paper literature strategist agent applies the corpus-first flow during its literature search strategy phase. The PRE-SCREENED block sits inside the Search Strategy section of the Literature Search Report (per the agent's existing Output Format), immediately before the `Databases` line. The merged `final_included` set feeds the agent's downstream Annotated Bibliography, Literature Matrix, Research Gap Identification, and Recommended Sources by Paper Section outputs without altering their formats.
|
||||
|
||||
PR-B replaces this entire block with the full content (matching the bibliography_agent pattern), removes the LINT_STUB marker and Status line, and appends the strategist entry to `scripts/corpus_consumer_manifest.json`.
|
||||
When `literature_corpus[]` is non-empty and parses cleanly, the agent enters Step 1 pre-screening, and Step 2 search-fills-gap dispatches the external 4-Layer Progressive Strategy. When the corpus is absent, empty, or fails the minimal shape check, the agent runs its existing 4-Layer external-DB-only flow unchanged (Iron Rule 4 graceful fallback for the failure cases).
|
||||
|
||||
@@ -2,8 +2,8 @@
|
||||
name: deep-research
|
||||
description: "Universal deep research agent team. 13-agent pipeline for rigorous academic research on any topic. 7 modes: full research, quick brief, paper review, lit-review, fact-check, Socratic guided research dialogue, and systematic review with optional meta-analysis. Covers research question formulation, Socratic mentoring, methodology design, systematic literature search, source verification, cross-source synthesis, risk of bias assessment, meta-analysis, APA 7.0 report compilation, editorial review, devil's advocate challenges, ethics review, and post-research literature monitoring. Triggers on: research, deep research, literature review, systematic review, meta-analysis, PRISMA, evidence synthesis, fact-check, guide my research, help me think through, 研究, 深度研究, 文獻回顧, 文獻探討, 系統性回顧, 後設分析, 事實查核, 引導我的研究, 幫我釐清, 幫我想想, 我不確定要研究什麼, 研究方向, 研究主題."
|
||||
metadata:
|
||||
version: "2.9.1"
|
||||
last_updated: "2026-04-22"
|
||||
version: "2.9.2"
|
||||
last_updated: "2026-04-27"
|
||||
status: active
|
||||
data_access_level: raw
|
||||
task_type: open-ended
|
||||
@@ -476,8 +476,8 @@ deep-research (systematic-review) + academic-paper -> PRISMA systematic review p
|
||||
|
||||
| Item | Content |
|
||||
|------|---------|
|
||||
| Skill Version | 2.9.0 |
|
||||
| Last Updated | 2026-04-20 |
|
||||
| Skill Version | 2.9.2 |
|
||||
| Last Updated | 2026-04-27 |
|
||||
| Maintainer | Cheng-I Wu |
|
||||
| Dependent Skills | academic-paper v1.0+ (downstream) |
|
||||
|
||||
|
||||
+4
-2
@@ -96,7 +96,7 @@ The Material Passport `literature_corpus[]` field is populated by user-written a
|
||||
- `literature_corpus[]` entries are sorted by `citation_key`; rejections are sorted by `source`.
|
||||
- Adapter output size grows linearly with corpus size. A 500-entry Zotero library typically produces a passport of ~300 KB YAML. ARS consumers should lazy-load when the corpus is large.
|
||||
|
||||
### What v3.6.4 does NOT do
|
||||
### Ingestion-layer boundaries
|
||||
|
||||
- Does not ingest PDFs, extract text, or run OCR.
|
||||
- Does not call the Zotero Web API, Notion API, or any live service.
|
||||
@@ -106,4 +106,6 @@ These boundaries are deliberate and reflect the ARS data-layer decision: ARS is
|
||||
|
||||
### Consumer-side integration
|
||||
|
||||
As of v3.6.4, no ARS agent reads `literature_corpus[]`. The field is a defined input port only. Consumer-side integration (agents that actually USE the corpus for research planning, citation generation, etc.) is deferred to v3.6.5+.
|
||||
As of v3.6.5, two Phase 1 literature agents read `literature_corpus[]` via the **corpus-first, search-fills-gap** flow: `deep-research/agents/bibliography_agent.md` and `academic-paper/agents/literature_strategist_agent.md`. Both consumers follow the same five-step shared flow and four Iron Rules (Same criteria / No silent skip / No corpus mutation / Graceful fallback on parse failure). Search Strategy reports gain a PRE-SCREENED reproducibility block that enumerates included / excluded / skipped corpus entries with F3 zero-hit and F4 provenance reporting. Consumer integration is presence-based — auto-engages when the passport carries a non-empty `literature_corpus[]` and parses cleanly; parse failures fall back to external-DB-only flow with a `[CORPUS PARSE FAILURE]` surface.
|
||||
|
||||
See [`academic-pipeline/references/literature_corpus_consumers.md`](../academic-pipeline/references/literature_corpus_consumers.md) for the full consumer protocol. `citation_compliance_agent` corpus integration is deferred to v3.6.6+.
|
||||
|
||||
@@ -96,7 +96,7 @@ Material Passport 的 `literature_corpus[]` 欄位由**使用者自行撰寫的
|
||||
- `literature_corpus[]` 依 `citation_key` 排序；`rejection_log.rejected[]` 依 `source` 排序。
|
||||
- Adapter 輸出大小與語料庫大小線性成長。500 筆 Zotero 書目約產出 300 KB 的 passport YAML。大型語料庫建議 ARS 消費端採 lazy load。
|
||||
|
||||
### v3.6.4 **不做** 的事
|
||||
### 導入層邊界
|
||||
|
||||
- 不讀 PDF 內容、不做文字抽取、不跑 OCR。
|
||||
- 不呼叫 Zotero Web API、Notion API 或任何遠端服務。
|
||||
@@ -106,4 +106,6 @@ Material Passport 的 `literature_corpus[]` 欄位由**使用者自行撰寫的
|
||||
|
||||
### 消費端整合
|
||||
|
||||
v3.6.4 時沒有任何 ARS agent 讀取 `literature_corpus[]`。此欄位僅為 input port 定義，消費端整合（agent 實際使用語料做研究規劃 / citation 生成 / ...）留待 v3.6.5+。
|
||||
v3.6.5 起，Phase 1 兩個文獻 agent 透過 **corpus-first、search-fills-gap** 流程讀取 `literature_corpus[]`：`deep-research/agents/bibliography_agent.md` 與 `academic-paper/agents/literature_strategist_agent.md`。兩者走相同的五步流程與四條 Iron Rule（Same criteria / No silent skip / No corpus mutation / Graceful fallback on parse failure）。Search Strategy 報告新增 PRE-SCREENED 可重現區塊，列出已納入／排除／略過的 corpus entry，並含 F3 zero-hit 與 F4 provenance 報告。消費端啟動採 presence-based — passport 帶非空 `literature_corpus[]` 且解析成功時自動進入；解析失敗時 fallback 到 external-DB-only flow，並 surface `[CORPUS PARSE FAILURE]`。
|
||||
|
||||
完整 consumer 協定見 [`academic-pipeline/references/literature_corpus_consumers.md`](../academic-pipeline/references/literature_corpus_consumers.md)。`citation_compliance_agent` 的 corpus 整合留到 v3.6.6+。
|
||||
|
||||
@@ -61,7 +61,7 @@ def check_relative_markdown_links(rel_path: str) -> None:
|
||||
def check_mode_registry() -> None:
|
||||
rel_path = "MODE_REGISTRY.md"
|
||||
text = read(rel_path)
|
||||
expect_contains(rel_path, "Last updated: v3.6.4 (2026-04-25)")
|
||||
expect_contains(rel_path, "Last updated: v3.6.5 (2026-04-27)")
|
||||
for heading in (
|
||||
"## deep-research (7 modes)",
|
||||
"## academic-paper (10 modes)",
|
||||
@@ -75,7 +75,7 @@ def check_claude_md() -> None:
|
||||
rel_path = ".claude/CLAUDE.md"
|
||||
expect_contains(rel_path, "integrity check (Stage 2.5)")
|
||||
expect_contains(rel_path, "final integrity check (Stage 4.5)")
|
||||
expect_contains(rel_path, "**Suite version**: 3.6.4")
|
||||
expect_contains(rel_path, "**Suite version**: 3.6.5")
|
||||
for forbidden in (
|
||||
"6th independent reviewer",
|
||||
"Peer review gains 6th independent reviewer",
|
||||
@@ -136,8 +136,9 @@ def check_readme_sections() -> None:
|
||||
rel_path = "README.md"
|
||||
text = read(rel_path)
|
||||
|
||||
expect_contains(rel_path, "version-v3.6.4-blue")
|
||||
expect_contains(rel_path, "releases/tag/v3.6.4")
|
||||
expect_contains(rel_path, "version-v3.6.5-blue")
|
||||
expect_contains(rel_path, "releases/tag/v3.6.5")
|
||||
expect_contains(rel_path, "### v3.6.5 (2026-04-27)")
|
||||
expect_contains(rel_path, "### v3.6.4 (2026-04-25)")
|
||||
expect_contains(rel_path, "### v3.6.3 (2026-04-23)")
|
||||
expect_contains(rel_path, "### v3.6.2 (2026-04-23)")
|
||||
@@ -199,8 +200,9 @@ def check_readme_zh_sections() -> None:
|
||||
rel_path = "README.zh-TW.md"
|
||||
text = read(rel_path)
|
||||
|
||||
expect_contains(rel_path, "version-v3.6.4-blue")
|
||||
expect_contains(rel_path, "releases/tag/v3.6.4")
|
||||
expect_contains(rel_path, "version-v3.6.5-blue")
|
||||
expect_contains(rel_path, "releases/tag/v3.6.5")
|
||||
expect_contains(rel_path, "### v3.6.5（2026-04-27）")
|
||||
expect_contains(rel_path, "### v3.6.4（2026-04-25）")
|
||||
expect_contains(rel_path, "### v3.6.3（2026-04-23）")
|
||||
expect_contains(rel_path, "### v3.6.2（2026-04-23）")
|
||||
|
||||
@@ -1,5 +1,5 @@
|
||||
{
|
||||
"$schema_note": "Manifest of supported literature_corpus[] consumer agents. PR-A pre-release state: single entry. PR-B release state: both entries.",
|
||||
"$schema_note": "Manifest of supported literature_corpus[] consumer agents. v3.6.5 release state: bibliography_agent (deep-research) + literature_strategist_agent (academic-paper).",
|
||||
"supported_consumers": [
|
||||
{
|
||||
"agent_path": "deep-research/agents/bibliography_agent.md",
|
||||
@@ -7,6 +7,13 @@
|
||||
"skill": "deep-research",
|
||||
"since_version": "v3.6.5",
|
||||
"phase": "Phase 1"
|
||||
},
|
||||
{
|
||||
"agent_path": "academic-paper/agents/literature_strategist_agent.md",
|
||||
"agent_basename": "literature_strategist_agent",
|
||||
"skill": "academic-paper",
|
||||
"since_version": "v3.6.5",
|
||||
"phase": "Phase 1"
|
||||
}
|
||||
]
|
||||
}
|
||||
|
||||
@@ -575,7 +575,7 @@ The optional `literature_corpus[]` field is Schema 9's input port for user-owned
|
||||
|
||||
ARS does not produce these entries. User-written adapters read their own corpus source (Zotero, Obsidian, folder, Notion, etc.) and emit a passport with `literature_corpus[]` populated. Three reference adapters ship with v3.6.4 under [`scripts/adapters/`](../scripts/adapters/).
|
||||
|
||||
Consumer-side integration deferred to v3.6.5+ — `literature_corpus[]`-reading agents land progressively across PR-A (`bibliography_agent`) and PR-B (`literature_strategist_agent`); v3.6.5 release tag goes on PR-B. v3.6.4 defines the input port only.
|
||||
Consumer integration ships in v3.6.5: `bibliography_agent` (deep-research, Phase 1) and `literature_strategist_agent` (academic-paper, Phase 1) read `literature_corpus[]` via the corpus-first, search-fills-gap flow. See [`academic-pipeline/references/literature_corpus_consumers.md`](../academic-pipeline/references/literature_corpus_consumers.md) for the full consumer protocol, the four Iron Rules, and per-consumer reading instructions.
|
||||
|
||||
See [`academic-pipeline/references/adapters/overview.md`](../academic-pipeline/references/adapters/overview.md) for the adapter contract.
|
||||
|
||||
|
||||
Reference in New Issue
Block a user