fix(#3683): wire gated learnings extraction into completion, align copy path (#3810)

* test(#3683): failing-first rows for learnings source resolution and wiring pins

* fix(#3683): wire gated learnings extraction into completion, align copy path

* test(#3683): register the learnings suite in the docs-guard lane, drop unverified markers

* fix(#3683): close review findings — per-item parsing, readdir guards, docs paths

* fix(#3683): route phase enumeration through the locator seam, fix assertion targets

* fix(#3683): merge execute-phase ack into the 3003 fragment, fix fidelity targets

* fix(#3663): replace the spent execute-phase ack entry with the 3683 re-arm

* chore(#3683): backfill changeset pr number

---------

Co-authored-by: sim <sim@local>
This commit is contained in:
Tom Boucher
2026-08-24 09:23:01 -04:00
committed by GitHub
parent 31fcb833ec
commit 4b84be1da4
13 changed files with 280 additions and 29 deletions

View File

@@ -1197,7 +1197,7 @@ Extract reusable patterns, anti-patterns, and architectural decisions from compl
| `--format` | Output format: `markdown` (default), `json` |
**Prerequisites:** Phase has been executed (SUMMARY.md files exist)
**Produces:** `.planning/learnings/{phase}-LEARNINGS.md`
**Produces:** `.planning/phases/{phase-dir}/{padded-phase}-LEARNINGS.md`
**Extracts:**
- Architectural decisions and their rationale

View File

@@ -2562,6 +2562,8 @@ capture_thought({
Users who run a memory / knowledge-base MCP server (for example, ExoCortex-style servers, `claude-mem`, or `mem0`-style servers) can implement this tool name to have learnings routed into their knowledge base automatically with `project`, `phase`, and `source` metadata. Everyone else can use `/gsd-extract-learnings` without any extra setup — the `LEARNINGS.md` artifact is the feature.
With `features.global_learnings: true`, phase completion runs the extraction for the just-completed phase automatically and copies the artifact to the global store at `~/.gsd/knowledge/` (#3683) — extraction and copy failures never block completion. With the gate off (the default), extraction stays fully manual.
---
### 114. Context-Window-Aware Prompt Thinning

View File

@@ -918,7 +918,7 @@ GSD の保証付きでアドホックタスクを実行します。
| `--format` | 出力フォーマット: `markdown`(デフォルト)、`json` |
**前提条件:** フェーズが実行済みであること(SUMMARY.md ファイルが存在すること)
**生成物:** `.planning/learnings/{phase}-LEARNINGS.md`
**生成物:** `.planning/phases/{phase-dir}/{padded-phase}-LEARNINGS.md`
**抽出内容:**
- アーキテクチャ上の決定とその根拠

View File

@@ -924,7 +924,7 @@ GSD 보장을 통해 애드혹 작업을 실행합니다.
| `--format` | 출력 형식: `markdown` (기본값), `json` |
**전제 조건:** 단계가 실행됨 (SUMMARY.md 파일 존재)
**생성 결과:** `.planning/learnings/{phase}-LEARNINGS.md`
**생성 결과:** `.planning/phases/{phase-dir}/{padded-phase}-LEARNINGS.md`
**추출 내용:**
- 아키텍처 결정 및 근거

View File

@@ -921,7 +921,7 @@ Extrai padrões reutilizáveis, antipadrões e decisões arquiteturais do trabal
| `--format` | Formato de saída: `markdown` (padrão), `json` |
**Pré-requisitos:** A fase foi executada (arquivos SUMMARY.md existem)
**Produz:** `.planning/learnings/{phase}-LEARNINGS.md`
**Produz:** `.planning/phases/{phase-dir}/{padded-phase}-LEARNINGS.md`
**Extrai:**
- Decisões arquiteturais e sua justificativa

View File

@@ -918,7 +918,7 @@ ROADMAP.md 中阶段的 CRUD 操作 — 通过单一合并命令添加、插入
| `--format` | 输出格式:`markdown`(默认)、`json` |
**前提条件:** 阶段已被执行(SUMMARY.md 文件已存在)
**产出:** `.planning/learnings/{phase}-LEARNINGS.md`
**产出:** `.planning/phases/{phase-dir}/{padded-phase}-LEARNINGS.md`
**提取内容:**
- 架构决策及其依据