uctm 1.3.0 → 1.4.0

package/agents/ko/specifier.md

@@ -32,44 +32,43 @@ You are the **Specifier** — 사용자 요청을 요구사항으로 명세화
 
 ### 3-1. STARTUP — 참조 파일 즉시 읽기 (REQUIRED)
 
-| 파일 | 목적 |
-|------|------|
-| `.claude/agents/file-content-schema.md` | 파일 포맷 스키마 (PLAN.md, TASK, Requirement.md 포맷) |
-| `.claude/agents/shared-prompt-sections.md` | 공통 규칙 (TASK ID 패턴, WORK-LIST 규칙, log_work 함수) |
-| `.claude/agents/xml-schema.md` | XML 통신 포맷 (dispatch / task-result 구조) |
-| `.claude/agents/work-activity-log.md` | Activity Log 규칙 (log_work 함수, STAGE 테이블) |
+**REFERENCES_DIR 결정**: 입력에서 `REFERENCES_DIR=...` 라인을 확인. 해당 절대 경로를 사용. 없으면 기본값 `.claude/agents` 사용.
+
+#### Reference Loading (ref-cache)
+
+1. 수신한 dispatch XML에 `<ref-cache>`가 있는지 확인한다
+2. 필요한 참조 파일별로:
+   - ref-cache에 있으면 → **파일 읽기 SKIP**, 캐시된 내용 사용
+   - ref-cache에 없으면 → `{REFERENCES_DIR}/{filename}.md`에서 읽고 ref-cache에 추가
+3. 작업 완료 시 병합된 `<ref-cache>`를 반환 dispatch XML에 포함한다
+4. **하위 호환성**: dispatch에 `<ref-cache>`가 없으면 기존 방식대로 모든 참조 파일을 읽는다 (기존 동작 유지)
+
+이 에이전트의 필수 참조 파일:
+
+| 파일 | ref-cache key |
+|------|---------------|
+| `{REFERENCES_DIR}/file-content-schema.md` | `file-content-schema` |
+| `{REFERENCES_DIR}/shared-prompt-sections.md` | `shared-prompt-sections` |
+| `{REFERENCES_DIR}/xml-schema.md` | `xml-schema` |
+| `{REFERENCES_DIR}/work-activity-log.md` | `work-activity-log` |
 
 ### 3-2. WORK ID 결정
 
 ```bash
-WORK_FS=$(ls -d works/WORK-* 2>/dev/null | grep -oP 'WORK-\K\d+' | sort -n | tail -1)
-WORK_FS=${WORK_FS:-0}
-WORK_LIST=$(grep -oP '^WORK-\K\d+' works/WORK-LIST.md 2>/dev/null | sort -n | tail -1)
-WORK_LIST=${WORK_LIST:-0}
-WORK_MAX=$(( WORK_FS > WORK_LIST ? WORK_FS : WORK_LIST ))
-echo "WORK-$(printf "%02d" $((WORK_MAX + 1)))"
-[ "$WORK_FS" != "$WORK_LIST" ] && echo "WARNING: FS=$WORK_FS, LIST=$WORK_LIST mismatch"
+LAST_ID=$(grep -oP 'LAST_WORK_ID: WORK-\K\d+' works/WORK-LIST.md 2>/dev/null)
+LAST_ID=${LAST_ID:-0}
+NEW_ID=$(printf "%02d" $((LAST_ID + 1)))
+echo "WORK-${NEW_ID}"
 ```
 
-IN_PROGRESS WORK 존재 시:
-> "현재 진행 중인 WORK-XX가 있습니다. 추가 TASK로 진행할까요, 새 WORK를 생성할까요?"
+IN_PROGRESS 또는 DONE WORK 존재 시:
+> "현재 진행 중(IN_PROGRESS)이거나 완료 대기(DONE) 상태인 WORK-XX가 있습니다. 추가 TASK로 진행할까요, 새 WORK를 생성할까요?"
 
 ### 3-3. 프로젝트 탐색 (Discovery)
 
-```bash
-# 1. CLAUDE.md 언어 설정 확인
-grep -oP '(?<=Language:\s?)[a-z]{2}' CLAUDE.md 2>/dev/null
-
-# 2. 프로젝트 정보
-cat CLAUDE.md 2>/dev/null || cat README.md 2>/dev/null
-
-# 3. 기술 스택
-cat package.json 2>/dev/null | head -50
-cat pyproject.toml 2>/dev/null | head -30
+→ 프로젝트 탐색 명령: `shared-prompt-sections.md` § 11 참조
 
-# 4. 구조 (겸임 시에만 — 단순 요구사항은 생략 가능)
-find . -maxdepth 3 -type f \( -name "*.md" -o -name "*.json" -o -name "*.toml" \) | grep -v node_modules | head -30
-```
+Note: 3단계(구조)는 Planner 겸임 시에만 실행 — 단순 요구사항은 생략 가능.
 
 ### 3-4. Requirement.md 작성
 
@@ -120,7 +119,7 @@ Requirement.md 작성 완료 후, **요구사항 자체의 복잡도**로 판단
 5.  PLAN.md 생성 (Execution-Mode: direct) → file-content-schema.md § 1
 6.  TASK-00.md 생성 → file-content-schema.md § 2
 7.  TASK-00_progress.md 생성 (Status: PENDING) → file-content-schema.md § 3
-8.  WORK-LIST.md IN_PROGRESS 추가
+8.  WORK-LIST.md IN_PROGRESS 행 추가 + LAST_WORK_ID 갱신
 9.  log_work PLAN "Requirement.md, PLAN.md, TASK-00.md 생성 완료 (겸임)"
 10. 생성된 산출물 요약을 사용자에게 제시하고 승인 요청 (요구사항 + 설계 통합 검토)
 11. dispatch XML 반환. **호출은 Main Claude가 수행한다.**
@@ -128,6 +127,7 @@ Requirement.md 작성 완료 후, **요구사항 자체의 복잡도**로 판단
 ```
 
 → dispatch XML 포맷: `xml-schema.md` § 1 참조 (to="builder", task="TASK-00", execution-mode="direct")
+→ 로드한 모든 참조 파일을 포함한 `<ref-cache>` 추가 (`xml-schema.md` § 6 참조)
 
 ### 3-7. Planner 위임 — 복잡 요구사항 (pipeline/full)
 
@@ -140,7 +140,7 @@ Requirement.md 작성 완료 후, **요구사항 자체의 복잡도**로 판단
 1.  mkdir works/WORK-NN/
 2.  log_work INIT "WORK-NN 생성 — Planner 위임"
 3.  Requirement.md 생성 → § 3-4
-4.  WORK-LIST.md IN_PROGRESS 추가
+4.  WORK-LIST.md IN_PROGRESS 행 추가 + LAST_WORK_ID 갱신
 5.  log_work REF "참조: ..."
 6.  생성된 Requirement.md 요약을 사용자에게 제시하고 기획 승인 요청
 7.  dispatch XML 반환. **호출은 Main Claude가 수행한다.**
@@ -148,25 +148,17 @@ Requirement.md 작성 완료 후, **요구사항 자체의 복잡도**로 판단
 ```
 
 → dispatch XML 포맷: `xml-schema.md` § 1 참조 (to="planner", execution-mode="full")
+→ 로드한 모든 참조 파일을 포함한 `<ref-cache>` 추가 (`xml-schema.md` § 6 참조)
 
 ### 3-8. Output Language Rule
 
 → 우선순위 규칙: `shared-prompt-sections.md` § 1 참조
+→ 로케일 감지: `shared-prompt-sections.md` § 9 참조
 
 specifier 고유 규칙:
 - dispatch `<context><language>` 필드로 resolved language 전달
 - Requirement.md, PLAN.md 모두 resolved language로 작성
 
-로케일 감지:
-```
-1. CLAUDE.md → "Language: xx" 확인
-2. 없으면 사용자에게 언어 질문
-3. 없으면 시스템 로케일 자동 감지
-   - Windows: powershell -c "[CultureInfo]::CurrentCulture.TwoLetterISOLanguageName"
-   - Linux/Mac: locale | grep LANG | grep -oP '[a-z]{2}' | head -1
-   - Fallback: "en"
-```
-
 ---
 
 ## 4. 제약사항 및 금지사항
@@ -191,9 +183,9 @@ specifier 고유 규칙:
 - "자동으로 진행" 명시 시에만 auto mode (현재 WORK 내에서만 유효)
 
 ### WORK-LIST.md 규칙
-→ `.claude/agents/shared-prompt-sections.md` § 8 참조
+→ `{REFERENCES_DIR}/shared-prompt-sections.md` § 8 참조
 
-- WORK 생성 시: `IN_PROGRESS` 추가
+- WORK 생성 시: `IN_PROGRESS` 행 추가 + `LAST_WORK_ID` 헤더 갱신
 
 ### 파일명 규칙
 - TASK 파일명: `TASK-XX.md` 형식

package/agents/ko/verifier.md

@@ -33,12 +33,25 @@ Builder가 완료한 TASK 결과물을 검증하여 빌드, 린트, 테스트, A
 
 ### 3-1. STARTUP — 참조 파일 즉시 읽기 (REQUIRED)
 
-| 파일 | 목적 |
-|------|------|
-| `.claude/agents/shared-prompt-sections.md` | 공통 규칙 |
-| `.claude/agents/xml-schema.md` | XML 통신 포맷 |
-| `.claude/agents/context-policy.md` | 슬라이딩 윈도우 규칙 |
-| `.claude/agents/work-activity-log.md` | Activity Log 규칙 (log_work 함수, STAGE 테이블) |
+**REFERENCES_DIR 결정**: 입력에서 `REFERENCES_DIR=...` 라인 또는 `<references-dir>` XML 요소를 확인. 해당 절대 경로를 사용. 없으면 기본값 `.claude/agents` 사용.
+
+#### Reference Loading (ref-cache)
+
+1. 수신한 dispatch XML에 `<ref-cache>`가 있는지 확인한다
+2. 필요한 참조 파일별로:
+   - ref-cache에 있으면 → **파일 읽기 SKIP**, 캐시된 내용 사용
+   - ref-cache에 없으면 → `{REFERENCES_DIR}/{filename}.md`에서 읽고 ref-cache에 추가
+3. 작업 완료 시 병합된 `<ref-cache>`를 반환 task-result XML에 포함한다
+4. **하위 호환성**: dispatch에 `<ref-cache>`가 없으면 기존 방식대로 모든 참조 파일을 읽는다 (기존 동작 유지)
+
+이 에이전트의 필수 참조 파일:
+
+| 파일 | ref-cache key |
+|------|---------------|
+| `{REFERENCES_DIR}/shared-prompt-sections.md` | `shared-prompt-sections` |
+| `{REFERENCES_DIR}/xml-schema.md` | `xml-schema` |
+| `{REFERENCES_DIR}/context-policy.md` | `context-policy` |
+| `{REFERENCES_DIR}/work-activity-log.md` | `work-activity-log` |
 
 ### 3-2. XML 입력 파싱
 
@@ -46,7 +59,7 @@ Builder가 완료한 TASK 결과물을 검증하여 빌드, 린트, 테스트, A
 
 ### 3-3. Step 0: Progress File Gate (CRITICAL)
 
-→ Gate 조건: `file-content-schema.md` § 3 참조 (파일 존재 + Status=COMPLETED + Files changed)
+→ Gate 조건: `shared-prompt-sections.md` § 12 참조
 
 CRITICAL 실패 시 즉시 중단. 이후 Step 진행 불가.
 

package/agents/ko/work-activity-log.md

@@ -10,19 +10,21 @@
 * 작업 진행 시 : 작업항목 및 작업내용 
 * 수행작업 완료 시 : 타 Agent에 전송한 프롬프트 메시지** Agent 시작 시 수신한 프롬프트 메시지 내용 (Required 필수)
 
-## log_work 함수
-
-```bash
-AGENT_NAME="SPECIFIER"  # 각 에이전트 파일에서 적절히 설정
-
-log_work() {
-  local WORK_ID="$1" AGENT="$2" STAGE="$3" DESC="$4"
-  mkdir -p "works/${WORK_ID}"
-  printf '[%s]_%s_%s_%s\n' \
-    "$(date '+%Y-%m-%dT%H:%M:%S')" "$AGENT" "$STAGE" "$DESC" \
-    >> "works/${WORK_ID}/work_${WORK_ID}.log"
-}
+## log_work 방법
+
+Activity log 기록에 **Bash를 사용하지 않는다**. `Write` 도구 (또는 `Edit` 도구로 추가)를 사용한다.
+
+각 로그 항목 포맷:
 ```
+[YYYY-MM-DDTHH:MM:SS]_AGENT_STAGE_설명
+```
+
+예시: INIT 단계 로그 기록 시, **Write** 도구로 `works/{WORK_ID}/work_{WORK_ID}.log`에 추가:
+```
+[2026-03-28T14:30:00]_SPECIFIER_INIT_WORK-09 생성 — Execution-Mode: direct
+```
+
+→ **Bash 명령 규칙: `shared-prompt-sections.md` § 13 참조**
 
 ---
 

package/agents/ko/xml-schema.md

@@ -8,6 +8,7 @@ uc-taskmanager 에이전트 간 XML 통신 포맷 정의.
 
 ```xml
 <dispatch to="{receiver}" work="{WORK_ID}" task="{TASK_ID}" execution-mode="{direct|pipeline|full}">
+  <references-dir>{참조 파일 디렉토리의 절대 경로}</references-dir>
   <context>
     <project>{project name}</project>
     <language>{lang_code}</language>

package/agents/{en/planner.md → planner.md}

@@ -36,34 +36,34 @@ WORK (unit of work)    — Goal unit of the user's request
 
 ### 3-1. STARTUP — Read Reference Files Immediately (REQUIRED)
 
-| File | Purpose |
-|------|---------|
-| `.claude/agents/file-content-schema.md` | File format schema (PLAN.md 7 fields, TASK format) |
-| `.claude/agents/shared-prompt-sections.md` | Common rules (TASK ID, WORK-LIST rules) |
-| `.claude/agents/work-activity-log.md` | Activity Log rules (log_work function, STAGE table) |
+**Resolve REFERENCES_DIR**: Check your input for `REFERENCES_DIR=...` line or `<references-dir>` XML element. Use that absolute path. If not provided, default to `.claude/agents`.
 
-### 3-2. Project Exploration (Discovery Process)
+#### Reference Loading (ref-cache)
 
-```bash
-# 1. Check existing WORKs
-ls -d works/WORK-* 2>/dev/null | sort -V | tail -1
+1. Check if `<ref-cache>` exists in the received dispatch XML
+2. For each required reference file:
+   - If present in ref-cache → **SKIP file read**, use cached content
+   - If absent from ref-cache → Read from `{REFERENCES_DIR}/{filename}.md` and add to ref-cache
+3. On task completion, include the merged `<ref-cache>` in the returned task-result XML
+4. **Backward compatibility**: If dispatch contains no `<ref-cache>`, read all reference files normally (existing behavior)
 
-# 2. Check CLAUDE.md language setting
-grep -oP '(?<=Language:\s?)[a-z]{2}' CLAUDE.md 2>/dev/null
+Required reference files for this agent:
 
-# 3. Project information
-cat CLAUDE.md 2>/dev/null || cat README.md 2>/dev/null
+| File | ref-cache key |
+|------|---------------|
+| `{REFERENCES_DIR}/file-content-schema.md` | `file-content-schema` |
+| `{REFERENCES_DIR}/shared-prompt-sections.md` | `shared-prompt-sections` |
+| `{REFERENCES_DIR}/work-activity-log.md` | `work-activity-log` |
 
-# 4. Tech stack
-cat package.json 2>/dev/null | head -50
-cat pyproject.toml 2>/dev/null | head -30
-cat Cargo.toml 2>/dev/null | head -20
-cat go.mod 2>/dev/null | head -10
+### 3-2. Project Exploration (Discovery Process)
 
-# 5. Structure
-find . -maxdepth 3 -type f \( -name "*.md" -o -name "*.json" -o -name "*.toml" \) | grep -v node_modules | head -30
+```bash
+# 1. Check existing WORKs
+ls -d works/WORK-* 2>/dev/null | sort -V | tail -1
 ```
 
+→ Discovery commands (steps 2–4): see `shared-prompt-sections.md` § 11
+
 ### 3-3. Requirement.md Analysis + WORK Directory Check
 
 Specifier has already created the WORK directory and written Requirement.md.
@@ -109,9 +109,11 @@ Record the determined mode in PLAN.md's `> Execution-Mode:` field.
 4. Completion report: "{WORK-ID} plan created. Start with `Run {WORK-ID} pipeline`."
 ```
 
+When returning scheduler or builder dispatch XML, include `<ref-cache>` with all reference files loaded (see `xml-schema.md` § 6).
+
 ### 3-6. Output Structure
 
-→ see `.claude/agents/file-content-schema.md` § 7
+→ see `{REFERENCES_DIR}/file-content-schema.md` § 7
 
 Creation responsibilities:
 - `PLAN.md`, `TASK-XX.md`, `TASK-XX_progress.md` (initial template) → Planner
@@ -121,7 +123,7 @@ Creation responsibilities:
 
 When creating TASK files, always create `TASK-XX_progress.md` template in the same directory.
 
-File formats: → `.claude/agents/file-content-schema.md` § 1 (PLAN.md), § 2 (TASK), § 3 (progress initial value)
+File formats: → `{REFERENCES_DIR}/file-content-schema.md` § 1 (PLAN.md), § 2 (TASK), § 3 (progress initial value)
 
 ### 3-7. MCP Tool Usage (Serena)
 
@@ -135,16 +137,7 @@ File formats: → `.claude/agents/file-content-schema.md` § 1 (PLAN.md), § 2 (
 ### 3-8. Output Language Rule
 
 → Priority rules: see `shared-prompt-sections.md` § 1
-
-Planner-specific locale detection:
-```
-1. CLAUDE.md → check "Language: xx"
-2. If not found, ask user for language
-3. If not found, auto-detect system locale
-   - Windows: powershell -c "[CultureInfo]::CurrentCulture.TwoLetterISOLanguageName"
-   - Linux/Mac: locale | grep LANG | grep -oP '[a-z]{2}' | head -1
-   - Fallback: "en"
-```
+→ Locale detection: see `shared-prompt-sections.md` § 9
 
 Record resolved language in PLAN.md `> Language:` field. Write all outputs in that language.
 

package/agents/{en/scheduler.md → scheduler.md}

@@ -36,13 +36,26 @@ You are the **Scheduler** — the WORK pipeline execution agent.
 
 ### 3-1. STARTUP — Read Reference Files Immediately (REQUIRED)
 
-| File | Purpose |
-|------|---------|
-| `.claude/agents/file-content-schema.md` | File format schema |
-| `.claude/agents/shared-prompt-sections.md` | Common rules |
-| `.claude/agents/xml-schema.md` | XML communication format |
-| `.claude/agents/context-policy.md` | Sliding Window rules |
-| `.claude/agents/work-activity-log.md` | Activity Log rules (log_work function, STAGE table) |
+**Resolve REFERENCES_DIR**: Check your input for `REFERENCES_DIR=...` line or `<references-dir>` XML element. Use that absolute path. If not provided, default to `.claude/agents`.
+
+#### Reference Loading (ref-cache)
+
+1. Check if `<ref-cache>` exists in the received dispatch XML
+2. For each required reference file:
+   - If present in ref-cache → **SKIP file read**, use cached content
+   - If absent from ref-cache → Read from `{REFERENCES_DIR}/{filename}.md` and add to ref-cache
+3. On task completion, include the merged `<ref-cache>` in the returned task-result XML
+4. **Backward compatibility**: If dispatch contains no `<ref-cache>`, read all reference files normally (existing behavior)
+
+Required reference files for this agent:
+
+| File | ref-cache key |
+|------|---------------|
+| `{REFERENCES_DIR}/file-content-schema.md` | `file-content-schema` |
+| `{REFERENCES_DIR}/shared-prompt-sections.md` | `shared-prompt-sections` |
+| `{REFERENCES_DIR}/xml-schema.md` | `xml-schema` |
+| `{REFERENCES_DIR}/context-policy.md` | `context-policy` |
+| `{REFERENCES_DIR}/work-activity-log.md` | `work-activity-log` |
 
 ### 3-2. WORK Identification and Initial Load
 
@@ -86,6 +99,7 @@ Process only TASKs within the WORK. Access to other WORKs prohibited.
 Send Pipeline Stage Callback before each stage starts (see § 3-6).
 
 → dispatch XML format: see `xml-schema.md` § 1 (to="builder", action="implement")
+→ Include `<ref-cache>` from previous task-result in dispatch XML (see `xml-schema.md` § 6 and `agent-flow.md` ref-cache Chain Propagation)
 
 Generate the dispatch XML below and return it. **Invocation is performed by Main Claude.**
 
@@ -113,6 +127,7 @@ FAIL → retry builder (max 3 times). 3 failures → pipeline halted.
 
 → dispatch XML format: see `xml-schema.md` § 1 (to="verifier", action="verify")
 → Sliding Window (Builder→Verifier): see `context-policy.md` Scheduler Dispatch section
+→ Include `<ref-cache>` from builder task-result in dispatch XML (see `xml-schema.md` § 6)
 
 Generate the dispatch XML below and return it. **Invocation is performed by Main Claude.**
 
@@ -121,6 +136,7 @@ Generate the dispatch XML below and return it. **Invocation is performed by Main
 → dispatch XML format: see `xml-schema.md` § 1 (to="committer", action="commit")
 → Sliding Window (Verifier FULL + Builder SUMMARY): see `context-policy.md` Scheduler Dispatch section
 → Inter-TASK Dependency Transfer: see `context-policy.md` Inter-TASK Dependency Transfer section
+→ Include `<ref-cache>` from verifier task-result in dispatch XML (see `xml-schema.md` § 6)
 
 Generate the dispatch XML below and return it. **Invocation is performed by Main Claude.**
 
@@ -132,7 +148,7 @@ Committer FAIL retry:
 
 ### 3-9. Progress Report
 
-Update PROGRESS.md after TASK completion (→ see `.claude/agents/file-content-schema.md` § 6) and output status:
+Update PROGRESS.md after TASK completion (→ see `{REFERENCES_DIR}/file-content-schema.md` § 6) and output status:
 
 ```
 ✅ TASK-XX completed — commit: {hash}
@@ -163,8 +179,8 @@ Multi-WORK status check:
 - Bypassing pipeline results in missing result.md → WORK completion recognition failure
 
 ### WORK-LIST.md Rules
-- Do not change WORK-LIST.md to COMPLETED — changed only at git push time
-- → see `.claude/agents/shared-prompt-sections.md` § 8
+- Do not modify WORK-LIST.md — archival is handled by committer
+- → see `{REFERENCES_DIR}/shared-prompt-sections.md` § 8
 
 ### Output Language Rule
 → see `shared-prompt-sections.md` § 1

package/agents/shared-prompt-sections.md

@@ -0,0 +1,252 @@
+# Shared Prompt Sections
+
+Common reusable sections. Each agent references these via `cache_control` markers.
+
+---
+
+## § 1. Output Language Rule
+
+```
+Priority: PLAN.md > Language: → CLAUDE.md ## Language → en (default)
+
+On dispatch: pass resolved language code in <context><language> field
+Section headers (##) are also written in the resolved language (refer to language mapping table)
+```
+
+---
+
+## § 2. Build and Lint Commands
+
+```bash
+# Auto-detect Build (execute only if script exists)
+if [ -f "package.json" ]; then
+  if node -e "const p=JSON.parse(require('fs').readFileSync('package.json','utf8')); process.exit(p.scripts&&p.scripts.build?0:1)" 2>/dev/null; then
+    npm run build 2>&1 || bun run build 2>&1 || yarn build 2>&1
+  fi
+elif [ -f "Cargo.toml" ]; then
+  cargo build 2>&1
+elif [ -f "go.mod" ]; then
+  go build ./... 2>&1
+elif [ -f "pyproject.toml" ] || [ -f "setup.py" ]; then
+  python -m py_compile $(find . -name "*.py" -not -path "*/venv/*" | head -20) 2>&1
+elif [ -f "Makefile" ]; then
+  make build 2>&1 || make 2>&1
+fi
+
+# Auto-detect Lint (execute only if script exists)
+if [ -f "package.json" ]; then
+  if node -e "const p=JSON.parse(require('fs').readFileSync('package.json','utf8')); process.exit(p.scripts&&p.scripts.lint?0:1)" 2>/dev/null; then
+    npm run lint 2>&1 || bun run lint 2>&1 || true
+  fi
+elif [ -f "pyproject.toml" ]; then
+  ruff check . 2>&1 || python -m flake8 . 2>&1 || true
+fi
+```
+
+- If build/lint scripts do not exist → **skip (treat as N/A)**.
+- On build/lint failure, always fix before reporting.
+
+---
+
+## § 3. WORK and TASK File Path Patterns
+
+```
+works/{WORK_ID}/
+  ├─ Requirement.md                 # Created by Specifier (mandatory)
+  ├─ PLAN.md
+  ├─ PROGRESS.md
+  ├─ TASK-00.md               # No WORK prefix
+  ├─ TASK-00_progress.md      # Separator: underscore
+  ├─ TASK-00_result.md        # Separator: underscore
+  └─ TASK-01.md ...
+```
+
+- WORK ID: `WORK-NN` (e.g., `WORK-03`)
+- TASK ID: `TASK-NN` (e.g., `TASK-00`) — WORK prefix must NOT be included
+
+---
+
+## § 4. File System Discovery Scripts
+
+```bash
+# Find latest WORK with incomplete TASKs
+for dir in $(ls -d works/WORK-* 2>/dev/null | sort -V -r); do
+  WORK_ID=$(basename $dir)
+  TOTAL=$(ls $dir/TASK-*.md 2>/dev/null | grep -v result | wc -l)
+  DONE=$(ls $dir/TASK-*_result.md 2>/dev/null | wc -l)
+  [ "$DONE" -lt "$TOTAL" ] && echo "$WORK_ID" && break
+done
+
+# List all WORKs
+ls -d works/WORK-* 2>/dev/null | sort -V
+
+# TASK completion status
+TOTAL=$(ls works/${WORK_ID}/TASK-*.md 2>/dev/null | grep -v result | wc -l)
+DONE=$(ls works/${WORK_ID}/TASK-*_result.md 2>/dev/null | wc -l)
+echo "$DONE / $TOTAL"
+```
+
+---
+
+## § 5. Task Result XML Format
+
+```xml
+<task-result work="{WORK_ID}" task="{TASK_ID}" agent="{agent}" status="{PASS|FAIL}">
+  <summary>{1-2 line summary}</summary>
+  <files-changed>
+    <file action="{created|modified|deleted}" path="{path}">{description}</file>
+  </files-changed>
+  <verification>
+    <check name="{type}" status="{PASS|FAIL|N/A}">{details}</check>
+  </verification>
+  <notes>{notes for next steps}</notes>
+</task-result>
+```
+
+---
+
+## § 7. PLAN.md Required Meta-Information — 7 Fields
+
+→ `{REFERENCES_DIR}/file-content-schema.md` § 1 reference
+
+| Field | Required | Description |
+|-------|----------|-------------|
+| `> Created:` | ✅ | YYYY-MM-DD |
+| `> Requirement:` | ✅ | `REQ-XXX` or user request text |
+| `> Execution-Mode:` | ✅ | `direct` / `pipeline` / `full` |
+| `> Project:` | ✅ | Project name |
+| `> Tech Stack:` | ✅ | Detected tech stack |
+| `> Language:` | ✅ | Language code (`ko`, `en`, etc.) |
+| `> Status:` | ✅ | Always starts as `PLANNED` |
+
+---
+
+## § 8. WORK-LIST.md Update Rules
+
+File: `works/WORK-LIST.md`
+
+**Format:**
+```
+LAST_WORK_ID: WORK-XX
+
+| WORK | 제목 | 상태 | 생성일 | 완료일 |
+|------|------|------|--------|--------|
+| WORK-NN | ... | IN_PROGRESS | YYYY-MM-DD | |
+| WORK-MM | ... | DONE | YYYY-MM-DD | YYYY-MM-DD |
+```
+
+| Status | Meaning | Trigger |
+|--------|---------|---------|
+| `IN_PROGRESS` | WORK is being executed | specifier creates WORK |
+| `DONE` | All TASKs completed, awaiting review/push | committer completes last TASK |
+| `COMPLETED` | Archived to _COMPLETED/ | push merge (Main Claude batch processes all DONE) |
+
+Rules:
+- `LAST_WORK_ID` header tracks the highest WORK ID ever created
+- **specifier**: on WORK creation, add IN_PROGRESS row + update `LAST_WORK_ID`
+- **committer**: when last TASK is completed, change `IN_PROGRESS` → `DONE` and fill completion date (do NOT move folder or remove row)
+- **Main Claude** (push procedure): move all DONE WORKs to `works/_COMPLETED/`, remove their rows from WORK-LIST.md
+
+---
+
+## § 9. Locale Detection
+
+```
+1. CLAUDE.md → check "Language: xx"
+2. If not found, ask user for language
+3. If not found, auto-detect system locale
+   - Windows: powershell -c "[CultureInfo]::CurrentCulture.TwoLetterISOLanguageName"
+   - Linux/Mac: locale | grep LANG | grep -oP '[a-z]{2}' | head -1
+   - Fallback: "en"
+```
+
+---
+
+## § 10. Callback Transmission Template
+
+→ **Bash command rules: see § 13** — each step below is a separate tool call.
+
+Replace `{CallbackType}` with the actual key name (e.g., `ProgressCallback`, `TaskCallback`).
+
+**Step 1.** Use `Grep` tool to find `{CallbackType}:` line in CLAUDE.md. If not found, skip callback entirely.
+
+**Step 2.** Use `Grep` tool to find `CallbackToken:` line in CLAUDE.md (optional).
+
+**Step 3.** Send callback with a single `curl` command:
+```bash
+curl -s -X POST "CALLBACK_URL" -H "Content-Type: application/json" -H "X-Runner-Api-Key: TOKEN" -d '{"workId":"WORK-01","taskId":"TASK-00",...}'
+```
+
+Agent-specific payload fields:
+- **ProgressCallback** (builder): `"status": "IN_PROGRESS"`, `"currentReasoning": "..."`
+- **TaskCallback** (committer): `"status": "SUCCESS"`, `"commitHash": "${COMMIT_HASH}"`
+
+---
+
+## § 11. Project Discovery
+
+```bash
+# 1. Check CLAUDE.md language setting
+grep -oP '(?<=Language:\s?)[a-z]{2}' CLAUDE.md 2>/dev/null
+
+# 2. Tech stack
+cat package.json 2>/dev/null | head -50
+cat pyproject.toml 2>/dev/null | head -30
+cat Cargo.toml 2>/dev/null | head -20
+cat go.mod 2>/dev/null | head -10
+
+# 3. Structure (when needed)
+find . -maxdepth 3 -type f \( -name "*.md" -o -name "*.json" -o -name "*.toml" \) | grep -v node_modules | head -30
+```
+
+---
+
+## § 12. Progress File Gate Check
+
+Gate conditions for `works/WORK-NN/TASK-XX_progress.md`:
+- File exists at the expected path
+- `Status: COMPLETED` line is present
+- `## Files Changed` section is present and non-empty
+
+On gate failure → return FAIL task-result immediately. Do not proceed to subsequent steps.
+
+---
+
+## § 13. Bash Command Rules
+
+Bash commands MUST follow these rules for permission compatibility.
+
+**MANDATORY:**
+- One simple command per Bash call — NO compound commands (`&&`, `||`, `;`, `|`)
+- NO `cd dir && command` — you are already in the project root
+- NO multi-line scripts — split into separate Bash calls
+- NO sub-shell expansions in arguments — e.g., `$(date ...)` inside `printf`
+- Use relative paths from project root (e.g., `works/WORK-01/`) — NO absolute paths
+- Use `git add file`, `git commit -m "msg"` — NO `git -C path` flag
+
+**For file operations, prefer dedicated tools over Bash:**
+- Read files → `Read` tool (NOT `cat`)
+- Write/append files → `Write` tool (NOT `echo >>` or `printf >>`)
+- Edit files → `Edit` tool (NOT `sed -i`)
+- Search files → `Grep` tool (NOT `grep`)
+- Find files → `Glob` tool (NOT `find`)
+
+**Activity log example:**
+```
+WRONG: printf '[%s]_%s\n' "$(date ...)" "INIT" >> work.log
+RIGHT: Use Write tool to append a line to the log file
+```
+
+**Git example:**
+```
+WRONG: cd /path/to/project && git add file && git commit -m "msg"
+RIGHT: git add file        (one call)
+       git commit -m "msg"  (next call)
+```
+
+---
+
+## Version
+
+- Created: 2026-03-10
+- Updated: 2026-03-28