adopt
稽核既有專案的產出物格式是否合規，產出編號的遷移計畫
中文翻譯
---
name: adopt
description: "既有專案（brownfield）稽核：現有的產出物真的能用嗎？產出編號的遷移計畫。和 /project-stage-detect 不同，它檢查的是合規性，不是有沒有存在。"
argument-hint: "[focus: full | gdds | adrs | stories | infra]"
user-invocable: true
allowed-tools: Read, Glob, Grep, Write, AskUserQuestion, Bash(bash "*/.claude/skills/adopt/../../hooks/yaml-helper.sh" resolve_config *), Bash(bash .claude/scripts/gdd-structure-check.sh *), Bash(bash .claude/scripts/migrate-v1-config.sh --dry-run*)
model: sonnet
---
!`bash "${CLAUDE_SKILL_DIR}/../../hooks/yaml-helper.sh" resolve_config --keys review_mode,automation,automation_always_ask,workflow`
上面已解析完成，直接使用。`/adopt` 在回報和寫入遷移狀態時，也會直接檢視 `project.yaml` 和舊版
設定檔；那種原始檢視是刻意的，和上面解析出來的值是分開的。
# Adopt：既有專案採用範本
這個 skill 會稽核既有專案的產出物，看它們是否**符合格式**，能不能配合範本的
skill 流程，然後產出排好優先順序的遷移計畫。
**這不是 `/project-stage-detect`。**
`/project-stage-detect` 回答的是：*有什麼東西存在？*
`/adopt` 回答的是：*現有的東西真的能配合範本的 skill 運作嗎？*
一個專案可以有 GDD、ADR 和 story，但如果這些產出物的內部格式不對，
每個對格式敏感的 skill 還是會默默失敗，或產出錯誤的結果。
**輸出：** `docs/adoption-plan-[date].md`，一份持續存在、可勾選的遷移計畫。
**參數模式：**
**稽核模式：** `$ARGUMENTS`（空白 = `full`）
- **沒有參數 / `full`**：完整稽核，涵蓋所有產出物類型
- **`gdds`**：只檢查 GDD 的格式合規
- **`adrs`**：只檢查 ADR 的格式合規
- **`stories`**：只檢查 story 的格式合規
- **`infra`**：只檢查基礎設施產出物的缺口（registry、manifest、sprint-status、stage.txt）
---
**自動化模式**：解析 `modes.automation`（`project.local.yaml` →
`project.yaml` → 預設 `collaborative`）。每次呼叫 `AskUserQuestion` 都要遵循 `.claude/docs/automation-modes.md`
（collaborative 一律詢問 · guided 只問重大事項 · autonomous 記錄後直接進行；
`automation_always_ask` 的類別一律會詢問）。
**`workflow`**（依 `.claude/docs/workflow-modes.md`）。它決定第 2 階段稽核的範圍和
第 3 階段的嚴重度：`full` 以完整結構稽核所有文件類型；`standard` 只稽核
必要的文件/章節（選用的章節只是資訊，不算缺口）；
`minimal` 只做 `design/game-brief.md` 的格式檢查，不期待有 GDD/ADR/UX。見第 2 階段。
## 第 1 階段：偵測專案狀態
**現在就讀 `references/1-detect.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 第 2 階段：格式稽核
**現在就讀 `references/2-format-audit.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 第 3 階段：為缺口分類並排優先順序
**現在就讀 `references/3-classify.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 第 4 階段：建立遷移計畫
**現在就讀 `references/4-migration-plan.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 第 5 階段：呈現摘要並詢問是否寫入
**現在就讀 `references/5-summary.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 第 6 階段：寫入採用計畫
**現在就讀 `references/6-write.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 第 7 階段：提供第一步行動
**現在就讀 `references/7-first-action.md` 並照著做**：這個階段的步驟在那個檔案裡，不在這裡。如果對話在這個階段被壓縮或恢復了，要重新讀一次。
## 協作協定
**適用於 `collaborative` 模式（預設）。** `guided` 和
`autonomous` 模式請見 `.claude/docs/automation-modes.md`，下面的規則
描述的是 collaborative 模式的要求，不是所有模式通用的行為。
1. **靜靜地讀** — 完整稽核做完，才呈現任何東西
2. **先給摘要** — 在詢問要不要寫入之前，先讓使用者看到範圍
3. **寫入前先問** — 建立採用計畫檔案之前，一定要先確認
4. **提供，不強迫** — 計畫只是建議，要修什麼、什麼時候修，由使用者決定
5. **一次一個行動** — 交出計畫之後，提供一個具體的下一步，
不要一次列出六件要同時做的事
6. **絕對不要重新產生既有的產出物** — 只補現有內容的缺口；
不要重寫已經有內容的 GDD、ADR 或 story英文原文
---
name: adopt
description: "Brownfield audit — do existing artifacts actually work? Numbered migration plan. Unlike /project-stage-detect, checks compliance not existence."
argument-hint: "[focus: full | gdds | adrs | stories | infra]"
user-invocable: true
allowed-tools: Read, Glob, Grep, Write, AskUserQuestion, Bash(bash "*/.claude/skills/adopt/../../hooks/yaml-helper.sh" resolve_config *), Bash(bash .claude/scripts/gdd-structure-check.sh *), Bash(bash .claude/scripts/migrate-v1-config.sh --dry-run*)
model: sonnet
---
!`bash "${CLAUDE_SKILL_DIR}/../../hooks/yaml-helper.sh" resolve_config --keys review_mode,automation,automation_always_ask,workflow`
Resolved above — use as-is. `/adopt` also inspects `project.yaml` and the legacy
config files directly when reporting and writing migration state; that raw
inspection is deliberate and separate from the resolved values above.
# Adopt — Brownfield Template Adoption
This skill audits an existing project's artifacts for **format compliance** with
the template's skill pipeline, then produces a prioritised migration plan.
**This is not `/project-stage-detect`.**
`/project-stage-detect` answers: *what exists?*
`/adopt` answers: *will what exists actually work with the template's skills?*
A project can have GDDs, ADRs, and stories — and every format-sensitive skill
will still fail silently or produce wrong results if those artifacts are in the
wrong internal format.
**Output:** `docs/adoption-plan-[date].md` — a persistent, checkable migration plan.
**Argument modes:**
**Audit mode:** `$ARGUMENTS` (blank = `full`)
- **No argument / `full`**: Complete audit — all artifact types
- **`gdds`**: GDD format compliance only
- **`adrs`**: ADR format compliance only
- **`stories`**: Story format compliance only
- **`infra`**: Infrastructure artifact gaps only (registry, manifest, sprint-status, stage.txt)
---
**Automation mode**: Resolve `modes.automation` (`project.local.yaml` →
`project.yaml` → default `collaborative`). Every `AskUserQuestion` call follows `.claude/docs/automation-modes.md`
(collaborative asks always · guided major-only · autonomous logs and proceeds;
`automation_always_ask` categories always prompt).
**`workflow`** (per `.claude/docs/workflow-modes.md`). It scopes the Phase 2 audit and
Phase 3 severity: `full` audits all doc types at full structure; `standard` audits
only the required docs/sections (optional sections are informational, not gaps);
`minimal` is a `design/game-brief.md` format check — GDDs/ADRs/UX are not expected. See Phase 2.
## Phase 1: Detect Project State
**Read `references/1-detect.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Phase 2: Format Audit
**Read `references/2-format-audit.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Phase 3: Classify and Prioritise Gaps
**Read `references/3-classify.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Phase 4: Build the Migration Plan
**Read `references/4-migration-plan.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Phase 5: Present Summary and Ask to Write
**Read `references/5-summary.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Phase 6: Write the Adoption Plan
**Read `references/6-write.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Phase 7: Offer First Action
**Read `references/7-first-action.md` now and follow it** — this phase's steps are in that file, not here. Read it again if the conversation was compacted or resumed during this phase.
## Collaborative Protocol
**Applies in `collaborative` mode (the default).** For `guided` and
`autonomous` modes, see `.claude/docs/automation-modes.md` — the rules below
describe what collaborative mode requires, not universal behavior.
1. **Read silently** — complete the full audit before presenting anything
2. **Show the summary first** — let the user see scope before asking to write
3. **Ask before writing** — always confirm before creating the adoption plan file
4. **Offer, don't force** — the plan is advisory; the user decides what to fix and when
5. **One action at a time** — after handing off the plan, offer one specific next step,
not a list of six things to do simultaneously
6. **Never regenerate existing artifacts** — only fill gaps in what exists;
do not rewrite GDDs, ADRs, or stories that already have content