@anhth2/spec-driven-dev-plugin 0.10.0 → 0.12.0

package/commands/fix-bug.md

@@ -84,7 +84,7 @@ Wait for explicit "Y" or "N" from the user before continuing.
 - "N" → stop and ask what the user wants to change.
 
 
-*Note: For this command, the target in Step 1 is a ticket ID (e.g., `PROJ-123`) or bug description from `$ARGUMENTS`. There is no file to resolve — proceed to context loading.*
+*Note: For this command, the target in Step 1 is a ticket ID (e.g., `PROJ-123`), a **filed bug `{BUG-ID}`** (a `{paths.bug_reports_dir}/{BUG-ID}.md` from `/report-bug` — e.g. a QC product-gap), or a bug description from `$ARGUMENTS`. If a `{BUG-ID}` is given, resolve & read that report for spec context; otherwise proceed to context loading.*
 
 ## Context
 # Context Loader — Load All Project Context
@@ -125,6 +125,8 @@ Read `.agent/project-context.yaml`. Extract and store:
 - `paths.specs_dir` → BDD specs root
 - `paths.prd_dir` → PRD documents root
 - `paths.refinement_dir` → findings/review output dir
+- `paths.qc_dir` → QC automation artifacts root (visible top-level, one subfolder per UC: `{qc_dir}/{UC-ID}/`)
+- `paths.qc_skills_dir` → where qc-* commands load QC skills from (default bundled `.agent/skills/qc`; override to the QC team's own repo/submodule so framework upgrade won't overwrite them)
 - `paths.product_definitions_dir` → product definitions root
 - `paths.domain_knowledge_dir` → domain knowledge root
 - `paths.business_dictionary` → path to business-dictionary.md
@@ -137,6 +139,8 @@ If `paths` section is absent, use these defaults:
 - `specs_dir` = `specs/bdd`
 - `prd_dir` = `specs/prd`
 - `refinement_dir` = `.agent/review`
+- `qc_dir` = `docs`
+- `qc_skills_dir` = `.agent/skills/qc`
 - `product_definitions_dir` = `specs/product-definition`
 - `domain_knowledge_dir` = `specs/domain-knowledge`
 - `business_dictionary` = `specs/domain-knowledge/business-dictionary.md`
@@ -181,6 +185,7 @@ If `services` section is present:
 - Override `paths.core_entities` → `{spec_source}/specs/domain-knowledge/core-entities.md`
 - Override `paths.bug_reports_dir` → `{spec_source}/feedback/bug-reports`
 - Override `paths.bdd_proposals_dir` → `{spec_source}/feedback/bdd-proposals`
+- Override `paths.prd_change_requests_dir` → `{spec_source}/feedback/prd-change-requests`
 
 > **Why under `spec_source`:** PRD, design-spec, domain knowledge, the **API contract (tech-docs)**, and tester feedback are all **cross-team artifacts** — they must live in the **shared spec repo** so every umbrella (FE/App/BE) reads the same source via `/sync`. Tech-docs specifically: BE authors the tech-design (API contract), commits + pushes it into the spec submodule (2-layer commit), and FE/App pull it on their next `/sync` to wire the real API in `/generate-code --phase=integration`. In single-service mode (no `spec_source`), these default under the repo root — still shared, same repo.
 
@@ -221,19 +226,41 @@ If the file does not exist → skip silently.
 
 ---
 
-## Step 3 — [CRITICAL] Load CLAUDE.md
+## Step 3 — [CRITICAL] Load CLAUDE.md (layered: root + service overlay)
 
 *This is the highest-priority context — it defines HOW to write code and documents for this project.*
 
-Read `CLAUDE.md`. Extract and store:
+CLAUDE.md is loaded in **two layers** so umbrella-wide rules and service-specific
+architecture/coding standards compose correctly. The agent always sits at the umbrella
+root, but the implementation code lives in a service submodule with its OWN stack,
+architecture, and conventions — so the service's CLAUDE.md must win for code generation.
+
+**Layer 1 — [BASE] Root CLAUDE.md (umbrella-wide).**
+Read `CLAUDE.md` at the repo root. Treat its contents as the **shared baseline** for the
+whole umbrella — git conventions, data-protection posture, cross-cutting rules, and (in
+single-service mode) the project's only architecture + coding standards.
+
+**Layer 2 — [OVERLAY] Service CLAUDE.md (umbrella mode only).**
+*Run only if `service_root` was set in Step 1.6 (i.e. a real service was routed to).*
+Read `{service_root}/CLAUDE.md`. This file defines the architecture + coding standards of
+the **actual stack being implemented** (e.g. `user-service` = java-spring, `web` = nextjs).
+Overlay it on top of Layer 1: **on any conflict, the service value WINS** for architecture,
+coding standards, and error handling. Layer-1 values that the service does not redefine
+(e.g. git conventions, banned patterns shared org-wide) remain in effect.
+
+From the **merged** result, extract and store:
 
 - **§1 Project Overview** → project name, language, framework, build/test commands, domains
-- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules
-- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns
-- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name
-- **§7 Git Conventions** → branch naming pattern, commit message format
+- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules — *service overlay wins*
+- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns — *service overlay wins*
+- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name — *service overlay wins*
+- **§7 Git Conventions** → branch naming pattern, commit message format — *root baseline unless service redefines*
 
-If `CLAUDE.md` does not exist → note it as missing and continue with project-context.yaml data only.
+**Resolution rules:**
+- If both layers exist → merge as above; record `claude_md_source = root + {service_root}`.
+- If only the service overlay exists (no root CLAUDE.md) → use the service file alone; `claude_md_source = {service_root}`.
+- If `service_root` is set but `{service_root}/CLAUDE.md` is **missing** → fall back to root CLAUDE.md and flag ⚠️ in the Step 7 recap (the service has no architecture/coding-standards definition — code generation will use umbrella defaults, which may be the wrong stack).
+- If neither exists → note CLAUDE.md as missing and continue with project-context.yaml data only.
 
 ---
 
@@ -333,7 +360,8 @@ Output exactly this block:
 [CTX LOADED]
 Stack     : {language} / {framework} / {database}
 Platform  : {active_module} ({platform_type})
-Layers    : {layer order from CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+Layers    : {layer order from merged CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+CLAUDE.md : {root + {service_root} | {service_root} only | root only | ⚠️ service overlay MISSING — using root | missing}
 Ticket    : {ticket_prefix}-
 Dict      : {loaded — N canonical terms, M banned terms | missing}
 Entities  : {loaded — EntityA, EntityB, EntityC | missing}
@@ -364,8 +392,9 @@ Proceed to the next step of the calling command.
 ---
 
 ## Phase 1 — Gather Info
+If a `{BUG-ID}` was given: read `{paths.bug_reports_dir}/{BUG-ID}.md` for the spec context, AC violated, expected-vs-actual, and the suggested layer — use it as the bug details (no need to re-ask).
 If ticket: fetch details (or ask user to paste).
-If no ticket — CHECKPOINT:
+If no ticket / BUG-ID — CHECKPOINT:
 1. Where does the bug occur? (module, endpoint, flow)
 2. Steps to reproduce?
 3. Expected vs Actual?
@@ -438,7 +467,11 @@ Proceed? (Y/N)
 ```
 
 ## Phase 3 — Fix
+
+*Umbrella mode: the buggy code lives in the **service submodule** resolved at context-loader Step 1.6 — create the branch and run every git/build step from **inside** `{service_root}`. Single-service: omit the `cd`.*
+
 ```bash
+cd {service_root}                              # umbrella: the service submodule; single-service: omit
 git checkout -b fix/{TICKET_ID}-{description}
 ```
 Apply fix. Add trace annotation if file has `@trace.implements`:
@@ -455,13 +488,37 @@ Test: "Regression {TICKET_ID}: {bug description}"
 ```
 Run test. If fail → debug and fix (max 3 iterations).
 
-## Phase 5 — Build & Commit
+## Phase 5 — Build & Commit (2-layer push in umbrella mode)
 ```bash
-{conventions.build_command}   # max 3 retries
+{conventions.build_command}   # max 3 retries — runs inside {service_root} in umbrella mode
+# Tầng 1 — push the fix branch in the service submodule (where the code lives):
 git add {files}
 git commit -m "fix({TICKET_ID}): {description}"
-git push -u origin fix/{TICKET_ID}-{slug}
+git push -u origin fix/{TICKET_ID}-{slug}      # then open a PR into the service's tracked branch
 ```
+> **Umbrella mode — Tầng 2 (bump umbrella pointer):** the umbrella records a *commit* of the service
+> submodule, not a branch. After the fix-branch PR **merges** into the service's tracked branch, bump
+> the pointer so teammates pulling the umbrella don't hit "commit not found":
+> ```bash
+> cd -                                          # back to umbrella root
+> git add {service_root} && git commit -m "chore: bump {service_root} pointer (fix {TICKET_ID})"
+> git push
+> ```
+> Single-service mode: no umbrella pointer — Tầng 1 is the whole push. Full rule: Sync & Update §4.4 (commit 2 tầng).
+
+## Phase 5.5 — Close the bug report (if fixing a filed `{BUG-ID}`)
+
+*Skip if the target was a plain ticket/description (no `{BUG-ID}` file).*
+
+After the fix is committed, update `{paths.bug_reports_dir}/{BUG-ID}.md`:
+- Set `State` → `🟡 Fixed` and append a short **Resolution** (root cause + commit/PR link).
+- It is **not** `Closed` yet — QC owns verification: when `/qc-run-test` re-runs and `qc_status`
+  for the linked SC flips to `pass`, it becomes `🟢 Closed` (and `qc_owner`/`qc_blocked_by` clear).
+- Commit the updated report to the spec repo (same 2-layer push as `/report-bug`) so the PO/PM
+  "waiting-on" view reflects it on `/sync`.
+
+This is the only write `/fix-bug` makes to the feedback area — it still fixes **code** only,
+never edits PRD/BDD (spec changes are PO/Dev per BUG_FLOW Case 2–4).
 
 ## Phase 6 — Offer to Record a Lesson (optional)
 
@@ -596,6 +653,13 @@ Suggest the logical next command based on workflow phase:
 | /generate-design-spec   | Designer review → Figma links confirmed → PO + Designer sign-off → `/generate-bdd {prd-file}` |
 | /generate-bdd           | `/review-context {feature-file}` to verify coverage |
 | /review-context (BDD)   | `/generate-tech-docs {UC-ID}` if APPROVED; regenerate if NEEDS_FIX |
+| /qc-analyze             | `/qc-plan {UC-ID}` (resolve 🔴 blocker gaps first) |
+| /qc-plan                | `/qc-design-test {UC-ID}`                     |
+| /qc-design-test         | `/qc-review {UC-ID}` (test-case review)       |
+| /qc-review (test-case)  | `/qc-run-test {UC-ID}` if APPROVED; fix TCs if NEEDS_FIX |
+| /qc-run-test            | `/qc-report {UC-ID}` then `/qc-review {UC-ID}` (script review) |
+| /qc-review (script)     | `/qc-report {UC-ID}` then create PR if APPROVED |
+| /qc-report              | `/validate-traces {UC-ID}` to refresh Living Docs (qc_status) |
 | /generate-tech-docs     | `/review-tech-docs {tech-design-file}`        |
 | /review-tech-docs       | `/generate-code {feature-file}` if APPROVED; fix doc if NEEDS_FIX |
 | /generate-code          | First gen → `/review-code {UC-ID}`; re-gen → `/dev-gen-test {UC-ID}` |
@@ -627,7 +691,8 @@ Next   : {suggested command with example arguments}
 Root Cause: {analysis}
 Changes: {list}
 ✅ Regression test added | ✅ Build: SUCCESS
+{🐞 BUG-{id} → State: Fixed (pushed) — Closed after /qc-run-test re-verify pass | if fixing a filed bug}
 {📝 Lesson L-NNN recorded (if captured)}
 Branch: fix/{TICKET_ID}-{slug}
-Next: Create PR and link to ticket.
+Next: Create PR and link to ticket. {QC: re-run /qc-run-test {UC-ID} to verify + close the bug | if applicable}
 ```

package/commands/fix-bug.tmpl

@@ -3,7 +3,7 @@
 ## Gate
 {{include:steps/gate.md}}
 
-*Note: For this command, the target in Step 1 is a ticket ID (e.g., `PROJ-123`) or bug description from `$ARGUMENTS`. There is no file to resolve — proceed to context loading.*
+*Note: For this command, the target in Step 1 is a ticket ID (e.g., `PROJ-123`), a **filed bug `{BUG-ID}`** (a `{paths.bug_reports_dir}/{BUG-ID}.md` from `/report-bug` — e.g. a QC product-gap), or a bug description from `$ARGUMENTS`. If a `{BUG-ID}` is given, resolve & read that report for spec context; otherwise proceed to context loading.*
 
 ## Context
 {{include:steps/context-loader.md}}
@@ -11,8 +11,9 @@
 ---
 
 ## Phase 1 — Gather Info
+If a `{BUG-ID}` was given: read `{paths.bug_reports_dir}/{BUG-ID}.md` for the spec context, AC violated, expected-vs-actual, and the suggested layer — use it as the bug details (no need to re-ask).
 If ticket: fetch details (or ask user to paste).
-If no ticket — CHECKPOINT:
+If no ticket / BUG-ID — CHECKPOINT:
 1. Where does the bug occur? (module, endpoint, flow)
 2. Steps to reproduce?
 3. Expected vs Actual?
@@ -85,7 +86,11 @@ Proceed? (Y/N)
 ```
 
 ## Phase 3 — Fix
+
+*Umbrella mode: the buggy code lives in the **service submodule** resolved at context-loader Step 1.6 — create the branch and run every git/build step from **inside** `{service_root}`. Single-service: omit the `cd`.*
+
 ```bash
+cd {service_root}                              # umbrella: the service submodule; single-service: omit
 git checkout -b fix/{TICKET_ID}-{description}
 ```
 Apply fix. Add trace annotation if file has `@trace.implements`:
@@ -102,13 +107,37 @@ Test: "Regression {TICKET_ID}: {bug description}"
 ```
 Run test. If fail → debug and fix (max 3 iterations).
 
-## Phase 5 — Build & Commit
+## Phase 5 — Build & Commit (2-layer push in umbrella mode)
 ```bash
-{conventions.build_command}   # max 3 retries
+{conventions.build_command}   # max 3 retries — runs inside {service_root} in umbrella mode
+# Tầng 1 — push the fix branch in the service submodule (where the code lives):
 git add {files}
 git commit -m "fix({TICKET_ID}): {description}"
-git push -u origin fix/{TICKET_ID}-{slug}
+git push -u origin fix/{TICKET_ID}-{slug}      # then open a PR into the service's tracked branch
 ```
+> **Umbrella mode — Tầng 2 (bump umbrella pointer):** the umbrella records a *commit* of the service
+> submodule, not a branch. After the fix-branch PR **merges** into the service's tracked branch, bump
+> the pointer so teammates pulling the umbrella don't hit "commit not found":
+> ```bash
+> cd -                                          # back to umbrella root
+> git add {service_root} && git commit -m "chore: bump {service_root} pointer (fix {TICKET_ID})"
+> git push
+> ```
+> Single-service mode: no umbrella pointer — Tầng 1 is the whole push. Full rule: Sync & Update §4.4 (commit 2 tầng).
+
+## Phase 5.5 — Close the bug report (if fixing a filed `{BUG-ID}`)
+
+*Skip if the target was a plain ticket/description (no `{BUG-ID}` file).*
+
+After the fix is committed, update `{paths.bug_reports_dir}/{BUG-ID}.md`:
+- Set `State` → `🟡 Fixed` and append a short **Resolution** (root cause + commit/PR link).
+- It is **not** `Closed` yet — QC owns verification: when `/qc-run-test` re-runs and `qc_status`
+  for the linked SC flips to `pass`, it becomes `🟢 Closed` (and `qc_owner`/`qc_blocked_by` clear).
+- Commit the updated report to the spec repo (same 2-layer push as `/report-bug`) so the PO/PM
+  "waiting-on" view reflects it on `/sync`.
+
+This is the only write `/fix-bug` makes to the feedback area — it still fixes **code** only,
+never edits PRD/BDD (spec changes are PO/Dev per BUG_FLOW Case 2–4).
 
 ## Phase 6 — Offer to Record a Lesson (optional)
 
@@ -135,7 +164,8 @@ If `Y` → run the capture procedure below with `source=/fix-bug {TICKET_ID}`, a
 Root Cause: {analysis}
 Changes: {list}
 ✅ Regression test added | ✅ Build: SUCCESS
+{🐞 BUG-{id} → State: Fixed (pushed) — Closed after /qc-run-test re-verify pass | if fixing a filed bug}
 {📝 Lesson L-NNN recorded (if captured)}
 Branch: fix/{TICKET_ID}-{slug}
-Next: Create PR and link to ticket.
+Next: Create PR and link to ticket. {QC: re-run /qc-run-test {UC-ID} to verify + close the bug | if applicable}
 ```

package/commands/generate-bdd.md

@@ -123,6 +123,8 @@ Read `.agent/project-context.yaml`. Extract and store:
 - `paths.specs_dir` → BDD specs root
 - `paths.prd_dir` → PRD documents root
 - `paths.refinement_dir` → findings/review output dir
+- `paths.qc_dir` → QC automation artifacts root (visible top-level, one subfolder per UC: `{qc_dir}/{UC-ID}/`)
+- `paths.qc_skills_dir` → where qc-* commands load QC skills from (default bundled `.agent/skills/qc`; override to the QC team's own repo/submodule so framework upgrade won't overwrite them)
 - `paths.product_definitions_dir` → product definitions root
 - `paths.domain_knowledge_dir` → domain knowledge root
 - `paths.business_dictionary` → path to business-dictionary.md
@@ -135,6 +137,8 @@ If `paths` section is absent, use these defaults:
 - `specs_dir` = `specs/bdd`
 - `prd_dir` = `specs/prd`
 - `refinement_dir` = `.agent/review`
+- `qc_dir` = `docs`
+- `qc_skills_dir` = `.agent/skills/qc`
 - `product_definitions_dir` = `specs/product-definition`
 - `domain_knowledge_dir` = `specs/domain-knowledge`
 - `business_dictionary` = `specs/domain-knowledge/business-dictionary.md`
@@ -179,6 +183,7 @@ If `services` section is present:
 - Override `paths.core_entities` → `{spec_source}/specs/domain-knowledge/core-entities.md`
 - Override `paths.bug_reports_dir` → `{spec_source}/feedback/bug-reports`
 - Override `paths.bdd_proposals_dir` → `{spec_source}/feedback/bdd-proposals`
+- Override `paths.prd_change_requests_dir` → `{spec_source}/feedback/prd-change-requests`
 
 > **Why under `spec_source`:** PRD, design-spec, domain knowledge, the **API contract (tech-docs)**, and tester feedback are all **cross-team artifacts** — they must live in the **shared spec repo** so every umbrella (FE/App/BE) reads the same source via `/sync`. Tech-docs specifically: BE authors the tech-design (API contract), commits + pushes it into the spec submodule (2-layer commit), and FE/App pull it on their next `/sync` to wire the real API in `/generate-code --phase=integration`. In single-service mode (no `spec_source`), these default under the repo root — still shared, same repo.
 
@@ -219,19 +224,41 @@ If the file does not exist → skip silently.
 
 ---
 
-## Step 3 — [CRITICAL] Load CLAUDE.md
+## Step 3 — [CRITICAL] Load CLAUDE.md (layered: root + service overlay)
 
 *This is the highest-priority context — it defines HOW to write code and documents for this project.*
 
-Read `CLAUDE.md`. Extract and store:
+CLAUDE.md is loaded in **two layers** so umbrella-wide rules and service-specific
+architecture/coding standards compose correctly. The agent always sits at the umbrella
+root, but the implementation code lives in a service submodule with its OWN stack,
+architecture, and conventions — so the service's CLAUDE.md must win for code generation.
+
+**Layer 1 — [BASE] Root CLAUDE.md (umbrella-wide).**
+Read `CLAUDE.md` at the repo root. Treat its contents as the **shared baseline** for the
+whole umbrella — git conventions, data-protection posture, cross-cutting rules, and (in
+single-service mode) the project's only architecture + coding standards.
+
+**Layer 2 — [OVERLAY] Service CLAUDE.md (umbrella mode only).**
+*Run only if `service_root` was set in Step 1.6 (i.e. a real service was routed to).*
+Read `{service_root}/CLAUDE.md`. This file defines the architecture + coding standards of
+the **actual stack being implemented** (e.g. `user-service` = java-spring, `web` = nextjs).
+Overlay it on top of Layer 1: **on any conflict, the service value WINS** for architecture,
+coding standards, and error handling. Layer-1 values that the service does not redefine
+(e.g. git conventions, banned patterns shared org-wide) remain in effect.
+
+From the **merged** result, extract and store:
 
 - **§1 Project Overview** → project name, language, framework, build/test commands, domains
-- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules
-- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns
-- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name
-- **§7 Git Conventions** → branch naming pattern, commit message format
+- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules — *service overlay wins*
+- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns — *service overlay wins*
+- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name — *service overlay wins*
+- **§7 Git Conventions** → branch naming pattern, commit message format — *root baseline unless service redefines*
 
-If `CLAUDE.md` does not exist → note it as missing and continue with project-context.yaml data only.
+**Resolution rules:**
+- If both layers exist → merge as above; record `claude_md_source = root + {service_root}`.
+- If only the service overlay exists (no root CLAUDE.md) → use the service file alone; `claude_md_source = {service_root}`.
+- If `service_root` is set but `{service_root}/CLAUDE.md` is **missing** → fall back to root CLAUDE.md and flag ⚠️ in the Step 7 recap (the service has no architecture/coding-standards definition — code generation will use umbrella defaults, which may be the wrong stack).
+- If neither exists → note CLAUDE.md as missing and continue with project-context.yaml data only.
 
 ---
 
@@ -331,7 +358,8 @@ Output exactly this block:
 [CTX LOADED]
 Stack     : {language} / {framework} / {database}
 Platform  : {active_module} ({platform_type})
-Layers    : {layer order from CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+Layers    : {layer order from merged CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+CLAUDE.md : {root + {service_root} | {service_root} only | root only | ⚠️ service overlay MISSING — using root | missing}
 Ticket    : {ticket_prefix}-
 Dict      : {loaded — N canonical terms, M banned terms | missing}
 Entities  : {loaded — EntityA, EntityB, EntityC | missing}
@@ -811,7 +839,7 @@ After generating all `.feature` files, create or update `{paths.trace_dir}/{UC-I
 
 **TSV columns (tab-separated, one header row + one data row per scenario):**
 ```
-sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tdev_selftest\tdev_selftest_at\tprd_version\tbdd_version\ttech_doc_revision\tprd_status\tuc_status\tfe_phase\tstatus\tlast_updated
+sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tdev_selftest\tdev_selftest_at\tqc_status\tqc_run_at\tqc_owner\tqc_blocked_by\tprd_version\tbdd_version\ttech_doc_revision\tprd_status\tuc_status\tfe_phase\tstatus\tlast_updated
 ```
 
 **Rules:**
@@ -819,7 +847,7 @@ sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tde
 - If file exists (re-generation) → for each SC in the new `.feature`:
   - SC already in `.tsv` AND `spec_ver` is unchanged → update only: `sc_title`, `prd_version`, `bdd_version`, `prd_status`, `uc_status`, `last_updated`. Leave all other columns unchanged.
   - SC already in `.tsv` AND `spec_ver` changed (scenario was modified) → update: `sc_title`, `spec_ver`, `prd_version`, `bdd_version`, `prd_status`, `uc_status`, `last_updated` AND set `status = DRIFT` immediately (so the TSV reflects drift without waiting for `/validate-traces`). Leave `gen_ver`, `implemented_by`, `test_count`, `test_classes`, `tech_doc_revision` unchanged.
-  - SC is new (added in this re-gen) → append new row with `gen_ver`, `implemented_by`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`, `tech_doc_revision` all set to `—`.
+  - SC is new (added in this re-gen) → append new row with `gen_ver`, `implemented_by`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`, `qc_status`, `qc_run_at`, `qc_owner`, `qc_blocked_by`, `tech_doc_revision` all set to `—`.
   - SC no longer in `.feature` (removed) → delete its row.
 
 **Values to write for each scenario:**
@@ -835,6 +863,10 @@ sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tde
 | `test_classes` | `—` |
 | `dev_selftest` | `—` (no tests run yet) |
 | `dev_selftest_at` | `—` |
+| `qc_status` | `—` (official QC automation result — set by `/qc-run-test`) |
+| `qc_run_at` | `—` |
+| `qc_owner` | `—` (who a non-passing SC waits on: `dev` / `po` — set by `/qc-run-test` + `/report-bug`) |
+| `qc_blocked_by` | `—` (linked `BUG-{id}` / `GAP-{id}` — set by `/qc-run-test` + `/report-bug`) |
 | `prd_version` | `@trace.prd_version` from `.feature` header |
 | `bdd_version` | `@trace.bdd_version` from `.feature` header |
 | `tech_doc_revision` | `—` |
@@ -903,6 +935,13 @@ Suggest the logical next command based on workflow phase:
 | /generate-design-spec   | Designer review → Figma links confirmed → PO + Designer sign-off → `/generate-bdd {prd-file}` |
 | /generate-bdd           | `/review-context {feature-file}` to verify coverage |
 | /review-context (BDD)   | `/generate-tech-docs {UC-ID}` if APPROVED; regenerate if NEEDS_FIX |
+| /qc-analyze             | `/qc-plan {UC-ID}` (resolve 🔴 blocker gaps first) |
+| /qc-plan                | `/qc-design-test {UC-ID}`                     |
+| /qc-design-test         | `/qc-review {UC-ID}` (test-case review)       |
+| /qc-review (test-case)  | `/qc-run-test {UC-ID}` if APPROVED; fix TCs if NEEDS_FIX |
+| /qc-run-test            | `/qc-report {UC-ID}` then `/qc-review {UC-ID}` (script review) |
+| /qc-review (script)     | `/qc-report {UC-ID}` then create PR if APPROVED |
+| /qc-report              | `/validate-traces {UC-ID}` to refresh Living Docs (qc_status) |
 | /generate-tech-docs     | `/review-tech-docs {tech-design-file}`        |
 | /review-tech-docs       | `/generate-code {feature-file}` if APPROVED; fix doc if NEEDS_FIX |
 | /generate-code          | First gen → `/review-code {UC-ID}`; re-gen → `/dev-gen-test {UC-ID}` |

package/commands/generate-bdd.tmpl

@@ -458,7 +458,7 @@ After generating all `.feature` files, create or update `{paths.trace_dir}/{UC-I
 
 **TSV columns (tab-separated, one header row + one data row per scenario):**
 ```
-sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tdev_selftest\tdev_selftest_at\tprd_version\tbdd_version\ttech_doc_revision\tprd_status\tuc_status\tfe_phase\tstatus\tlast_updated
+sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tdev_selftest\tdev_selftest_at\tqc_status\tqc_run_at\tqc_owner\tqc_blocked_by\tprd_version\tbdd_version\ttech_doc_revision\tprd_status\tuc_status\tfe_phase\tstatus\tlast_updated
 ```
 
 **Rules:**
@@ -466,7 +466,7 @@ sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tde
 - If file exists (re-generation) → for each SC in the new `.feature`:
   - SC already in `.tsv` AND `spec_ver` is unchanged → update only: `sc_title`, `prd_version`, `bdd_version`, `prd_status`, `uc_status`, `last_updated`. Leave all other columns unchanged.
   - SC already in `.tsv` AND `spec_ver` changed (scenario was modified) → update: `sc_title`, `spec_ver`, `prd_version`, `bdd_version`, `prd_status`, `uc_status`, `last_updated` AND set `status = DRIFT` immediately (so the TSV reflects drift without waiting for `/validate-traces`). Leave `gen_ver`, `implemented_by`, `test_count`, `test_classes`, `tech_doc_revision` unchanged.
-  - SC is new (added in this re-gen) → append new row with `gen_ver`, `implemented_by`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`, `tech_doc_revision` all set to `—`.
+  - SC is new (added in this re-gen) → append new row with `gen_ver`, `implemented_by`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`, `qc_status`, `qc_run_at`, `qc_owner`, `qc_blocked_by`, `tech_doc_revision` all set to `—`.
   - SC no longer in `.feature` (removed) → delete its row.
 
 **Values to write for each scenario:**
@@ -482,6 +482,10 @@ sc_id\tsc_title\tspec_ver\tgen_ver\timplemented_by\ttest_count\ttest_classes\tde
 | `test_classes` | `—` |
 | `dev_selftest` | `—` (no tests run yet) |
 | `dev_selftest_at` | `—` |
+| `qc_status` | `—` (official QC automation result — set by `/qc-run-test`) |
+| `qc_run_at` | `—` |
+| `qc_owner` | `—` (who a non-passing SC waits on: `dev` / `po` — set by `/qc-run-test` + `/report-bug`) |
+| `qc_blocked_by` | `—` (linked `BUG-{id}` / `GAP-{id}` — set by `/qc-run-test` + `/report-bug`) |
 | `prd_version` | `@trace.prd_version` from `.feature` header |
 | `bdd_version` | `@trace.bdd_version` from `.feature` header |
 | `tech_doc_revision` | `—` |

package/commands/generate-code.md

@@ -125,6 +125,8 @@ Read `.agent/project-context.yaml`. Extract and store:
 - `paths.specs_dir` → BDD specs root
 - `paths.prd_dir` → PRD documents root
 - `paths.refinement_dir` → findings/review output dir
+- `paths.qc_dir` → QC automation artifacts root (visible top-level, one subfolder per UC: `{qc_dir}/{UC-ID}/`)
+- `paths.qc_skills_dir` → where qc-* commands load QC skills from (default bundled `.agent/skills/qc`; override to the QC team's own repo/submodule so framework upgrade won't overwrite them)
 - `paths.product_definitions_dir` → product definitions root
 - `paths.domain_knowledge_dir` → domain knowledge root
 - `paths.business_dictionary` → path to business-dictionary.md
@@ -137,6 +139,8 @@ If `paths` section is absent, use these defaults:
 - `specs_dir` = `specs/bdd`
 - `prd_dir` = `specs/prd`
 - `refinement_dir` = `.agent/review`
+- `qc_dir` = `docs`
+- `qc_skills_dir` = `.agent/skills/qc`
 - `product_definitions_dir` = `specs/product-definition`
 - `domain_knowledge_dir` = `specs/domain-knowledge`
 - `business_dictionary` = `specs/domain-knowledge/business-dictionary.md`
@@ -181,6 +185,7 @@ If `services` section is present:
 - Override `paths.core_entities` → `{spec_source}/specs/domain-knowledge/core-entities.md`
 - Override `paths.bug_reports_dir` → `{spec_source}/feedback/bug-reports`
 - Override `paths.bdd_proposals_dir` → `{spec_source}/feedback/bdd-proposals`
+- Override `paths.prd_change_requests_dir` → `{spec_source}/feedback/prd-change-requests`
 
 > **Why under `spec_source`:** PRD, design-spec, domain knowledge, the **API contract (tech-docs)**, and tester feedback are all **cross-team artifacts** — they must live in the **shared spec repo** so every umbrella (FE/App/BE) reads the same source via `/sync`. Tech-docs specifically: BE authors the tech-design (API contract), commits + pushes it into the spec submodule (2-layer commit), and FE/App pull it on their next `/sync` to wire the real API in `/generate-code --phase=integration`. In single-service mode (no `spec_source`), these default under the repo root — still shared, same repo.
 
@@ -221,19 +226,41 @@ If the file does not exist → skip silently.
 
 ---
 
-## Step 3 — [CRITICAL] Load CLAUDE.md
+## Step 3 — [CRITICAL] Load CLAUDE.md (layered: root + service overlay)
 
 *This is the highest-priority context — it defines HOW to write code and documents for this project.*
 
-Read `CLAUDE.md`. Extract and store:
+CLAUDE.md is loaded in **two layers** so umbrella-wide rules and service-specific
+architecture/coding standards compose correctly. The agent always sits at the umbrella
+root, but the implementation code lives in a service submodule with its OWN stack,
+architecture, and conventions — so the service's CLAUDE.md must win for code generation.
+
+**Layer 1 — [BASE] Root CLAUDE.md (umbrella-wide).**
+Read `CLAUDE.md` at the repo root. Treat its contents as the **shared baseline** for the
+whole umbrella — git conventions, data-protection posture, cross-cutting rules, and (in
+single-service mode) the project's only architecture + coding standards.
+
+**Layer 2 — [OVERLAY] Service CLAUDE.md (umbrella mode only).**
+*Run only if `service_root` was set in Step 1.6 (i.e. a real service was routed to).*
+Read `{service_root}/CLAUDE.md`. This file defines the architecture + coding standards of
+the **actual stack being implemented** (e.g. `user-service` = java-spring, `web` = nextjs).
+Overlay it on top of Layer 1: **on any conflict, the service value WINS** for architecture,
+coding standards, and error handling. Layer-1 values that the service does not redefine
+(e.g. git conventions, banned patterns shared org-wide) remain in effect.
+
+From the **merged** result, extract and store:
 
 - **§1 Project Overview** → project name, language, framework, build/test commands, domains
-- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules
-- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns
-- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name
-- **§7 Git Conventions** → branch naming pattern, commit message format
+- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules — *service overlay wins*
+- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns — *service overlay wins*
+- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name — *service overlay wins*
+- **§7 Git Conventions** → branch naming pattern, commit message format — *root baseline unless service redefines*
 
-If `CLAUDE.md` does not exist → note it as missing and continue with project-context.yaml data only.
+**Resolution rules:**
+- If both layers exist → merge as above; record `claude_md_source = root + {service_root}`.
+- If only the service overlay exists (no root CLAUDE.md) → use the service file alone; `claude_md_source = {service_root}`.
+- If `service_root` is set but `{service_root}/CLAUDE.md` is **missing** → fall back to root CLAUDE.md and flag ⚠️ in the Step 7 recap (the service has no architecture/coding-standards definition — code generation will use umbrella defaults, which may be the wrong stack).
+- If neither exists → note CLAUDE.md as missing and continue with project-context.yaml data only.
 
 ---
 
@@ -333,7 +360,8 @@ Output exactly this block:
 [CTX LOADED]
 Stack     : {language} / {framework} / {database}
 Platform  : {active_module} ({platform_type})
-Layers    : {layer order from CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+Layers    : {layer order from merged CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+CLAUDE.md : {root + {service_root} | {service_root} only | root only | ⚠️ service overlay MISSING — using root | missing}
 Ticket    : {ticket_prefix}-
 Dict      : {loaded — N canonical terms, M banned terms | missing}
 Entities  : {loaded — EntityA, EntityB, EntityC | missing}
@@ -625,7 +653,7 @@ Update `{paths.trace_dir}/{UC-ID}.tsv` — for each implemented scenario, find t
 | `fe_phase` | `ui` if `--phase=ui` \| `integrated` if `--phase=integration` \| `—` if no phase flag |
 | `last_updated` | today `YYYY-MM-DD` |
 
-Leave all other columns (`sc_title`, `spec_ver`, `prd_version`, `prd_status`, `uc_status`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`) unchanged.
+Leave all other columns (`sc_title`, `spec_ver`, `prd_version`, `prd_status`, `uc_status`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`, `qc_status`, `qc_run_at`) unchanged.
 `status` is computed by `/validate-traces` — do not set here.
 
 ## Refresh Panel Mirror
@@ -693,6 +721,13 @@ Suggest the logical next command based on workflow phase:
 | /generate-design-spec   | Designer review → Figma links confirmed → PO + Designer sign-off → `/generate-bdd {prd-file}` |
 | /generate-bdd           | `/review-context {feature-file}` to verify coverage |
 | /review-context (BDD)   | `/generate-tech-docs {UC-ID}` if APPROVED; regenerate if NEEDS_FIX |
+| /qc-analyze             | `/qc-plan {UC-ID}` (resolve 🔴 blocker gaps first) |
+| /qc-plan                | `/qc-design-test {UC-ID}`                     |
+| /qc-design-test         | `/qc-review {UC-ID}` (test-case review)       |
+| /qc-review (test-case)  | `/qc-run-test {UC-ID}` if APPROVED; fix TCs if NEEDS_FIX |
+| /qc-run-test            | `/qc-report {UC-ID}` then `/qc-review {UC-ID}` (script review) |
+| /qc-review (script)     | `/qc-report {UC-ID}` then create PR if APPROVED |
+| /qc-report              | `/validate-traces {UC-ID}` to refresh Living Docs (qc_status) |
 | /generate-tech-docs     | `/review-tech-docs {tech-design-file}`        |
 | /review-tech-docs       | `/generate-code {feature-file}` if APPROVED; fix doc if NEEDS_FIX |
 | /generate-code          | First gen → `/review-code {UC-ID}`; re-gen → `/dev-gen-test {UC-ID}` |

package/commands/generate-code.tmpl

@@ -272,7 +272,7 @@ Update `{paths.trace_dir}/{UC-ID}.tsv` — for each implemented scenario, find t
 | `fe_phase` | `ui` if `--phase=ui` \| `integrated` if `--phase=integration` \| `—` if no phase flag |
 | `last_updated` | today `YYYY-MM-DD` |
 
-Leave all other columns (`sc_title`, `spec_ver`, `prd_version`, `prd_status`, `uc_status`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`) unchanged.
+Leave all other columns (`sc_title`, `spec_ver`, `prd_version`, `prd_status`, `uc_status`, `test_count`, `test_classes`, `dev_selftest`, `dev_selftest_at`, `qc_status`, `qc_run_at`) unchanged.
 `status` is computed by `/validate-traces` — do not set here.
 
 ## Refresh Panel Mirror

package/commands/generate-design-spec.md

@@ -125,6 +125,8 @@ Read `.agent/project-context.yaml`. Extract and store:
 - `paths.specs_dir` → BDD specs root
 - `paths.prd_dir` → PRD documents root
 - `paths.refinement_dir` → findings/review output dir
+- `paths.qc_dir` → QC automation artifacts root (visible top-level, one subfolder per UC: `{qc_dir}/{UC-ID}/`)
+- `paths.qc_skills_dir` → where qc-* commands load QC skills from (default bundled `.agent/skills/qc`; override to the QC team's own repo/submodule so framework upgrade won't overwrite them)
 - `paths.product_definitions_dir` → product definitions root
 - `paths.domain_knowledge_dir` → domain knowledge root
 - `paths.business_dictionary` → path to business-dictionary.md
@@ -137,6 +139,8 @@ If `paths` section is absent, use these defaults:
 - `specs_dir` = `specs/bdd`
 - `prd_dir` = `specs/prd`
 - `refinement_dir` = `.agent/review`
+- `qc_dir` = `docs`
+- `qc_skills_dir` = `.agent/skills/qc`
 - `product_definitions_dir` = `specs/product-definition`
 - `domain_knowledge_dir` = `specs/domain-knowledge`
 - `business_dictionary` = `specs/domain-knowledge/business-dictionary.md`
@@ -181,6 +185,7 @@ If `services` section is present:
 - Override `paths.core_entities` → `{spec_source}/specs/domain-knowledge/core-entities.md`
 - Override `paths.bug_reports_dir` → `{spec_source}/feedback/bug-reports`
 - Override `paths.bdd_proposals_dir` → `{spec_source}/feedback/bdd-proposals`
+- Override `paths.prd_change_requests_dir` → `{spec_source}/feedback/prd-change-requests`
 
 > **Why under `spec_source`:** PRD, design-spec, domain knowledge, the **API contract (tech-docs)**, and tester feedback are all **cross-team artifacts** — they must live in the **shared spec repo** so every umbrella (FE/App/BE) reads the same source via `/sync`. Tech-docs specifically: BE authors the tech-design (API contract), commits + pushes it into the spec submodule (2-layer commit), and FE/App pull it on their next `/sync` to wire the real API in `/generate-code --phase=integration`. In single-service mode (no `spec_source`), these default under the repo root — still shared, same repo.
 
@@ -221,19 +226,41 @@ If the file does not exist → skip silently.
 
 ---
 
-## Step 3 — [CRITICAL] Load CLAUDE.md
+## Step 3 — [CRITICAL] Load CLAUDE.md (layered: root + service overlay)
 
 *This is the highest-priority context — it defines HOW to write code and documents for this project.*
 
-Read `CLAUDE.md`. Extract and store:
+CLAUDE.md is loaded in **two layers** so umbrella-wide rules and service-specific
+architecture/coding standards compose correctly. The agent always sits at the umbrella
+root, but the implementation code lives in a service submodule with its OWN stack,
+architecture, and conventions — so the service's CLAUDE.md must win for code generation.
+
+**Layer 1 — [BASE] Root CLAUDE.md (umbrella-wide).**
+Read `CLAUDE.md` at the repo root. Treat its contents as the **shared baseline** for the
+whole umbrella — git conventions, data-protection posture, cross-cutting rules, and (in
+single-service mode) the project's only architecture + coding standards.
+
+**Layer 2 — [OVERLAY] Service CLAUDE.md (umbrella mode only).**
+*Run only if `service_root` was set in Step 1.6 (i.e. a real service was routed to).*
+Read `{service_root}/CLAUDE.md`. This file defines the architecture + coding standards of
+the **actual stack being implemented** (e.g. `user-service` = java-spring, `web` = nextjs).
+Overlay it on top of Layer 1: **on any conflict, the service value WINS** for architecture,
+coding standards, and error handling. Layer-1 values that the service does not redefine
+(e.g. git conventions, banned patterns shared org-wide) remain in effect.
+
+From the **merged** result, extract and store:
 
 - **§1 Project Overview** → project name, language, framework, build/test commands, domains
-- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules
-- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns
-- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name
-- **§7 Git Conventions** → branch naming pattern, commit message format
+- **§2 Architecture** → layer order (e.g., Controller → Facade → Service → Repository), architectural rules — *service overlay wins*
+- **§3 Coding Standards** → naming conventions (classes, methods), response wrapper type, forbidden patterns — *service overlay wins*
+- **§5 Error Handling** → exception types, HTTP status code mapping, not-found exception class name — *service overlay wins*
+- **§7 Git Conventions** → branch naming pattern, commit message format — *root baseline unless service redefines*
 
-If `CLAUDE.md` does not exist → note it as missing and continue with project-context.yaml data only.
+**Resolution rules:**
+- If both layers exist → merge as above; record `claude_md_source = root + {service_root}`.
+- If only the service overlay exists (no root CLAUDE.md) → use the service file alone; `claude_md_source = {service_root}`.
+- If `service_root` is set but `{service_root}/CLAUDE.md` is **missing** → fall back to root CLAUDE.md and flag ⚠️ in the Step 7 recap (the service has no architecture/coding-standards definition — code generation will use umbrella defaults, which may be the wrong stack).
+- If neither exists → note CLAUDE.md as missing and continue with project-context.yaml data only.
 
 ---
 
@@ -333,7 +360,8 @@ Output exactly this block:
 [CTX LOADED]
 Stack     : {language} / {framework} / {database}
 Platform  : {active_module} ({platform_type})
-Layers    : {layer order from CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+Layers    : {layer order from merged CLAUDE.md §2, e.g., Controller → Facade → Service → Repository}
+CLAUDE.md : {root + {service_root} | {service_root} only | root only | ⚠️ service overlay MISSING — using root | missing}
 Ticket    : {ticket_prefix}-
 Dict      : {loaded — N canonical terms, M banned terms | missing}
 Entities  : {loaded — EntityA, EntityB, EntityC | missing}
@@ -827,6 +855,13 @@ Suggest the logical next command based on workflow phase:
 | /generate-design-spec   | Designer review → Figma links confirmed → PO + Designer sign-off → `/generate-bdd {prd-file}` |
 | /generate-bdd           | `/review-context {feature-file}` to verify coverage |
 | /review-context (BDD)   | `/generate-tech-docs {UC-ID}` if APPROVED; regenerate if NEEDS_FIX |
+| /qc-analyze             | `/qc-plan {UC-ID}` (resolve 🔴 blocker gaps first) |
+| /qc-plan                | `/qc-design-test {UC-ID}`                     |
+| /qc-design-test         | `/qc-review {UC-ID}` (test-case review)       |
+| /qc-review (test-case)  | `/qc-run-test {UC-ID}` if APPROVED; fix TCs if NEEDS_FIX |
+| /qc-run-test            | `/qc-report {UC-ID}` then `/qc-review {UC-ID}` (script review) |
+| /qc-review (script)     | `/qc-report {UC-ID}` then create PR if APPROVED |
+| /qc-report              | `/validate-traces {UC-ID}` to refresh Living Docs (qc_status) |
 | /generate-tech-docs     | `/review-tech-docs {tech-design-file}`        |
 | /review-tech-docs       | `/generate-code {feature-file}` if APPROVED; fix doc if NEEDS_FIX |
 | /generate-code          | First gen → `/review-code {UC-ID}`; re-gen → `/dev-gen-test {UC-ID}` |