fix: enable dynamic inner_loop for parallel task execution across 6 team skills

- team-lifecycle-v4: executor inner_loop true→dynamic, add dynamicImplDispatch for PLAN-001 callback
- team-testing: executor inner_loop true→dynamic for comprehensive pipeline parallel TESTRUN
- team-quality-assurance: executor inner_loop true→dynamic for full mode parallel QARUN
- team-perf-opt: optimizer inner_loop true→dynamic for fan-out/independent parallel IMPL branches
- team-arch-opt: refactorer inner_loop true→dynamic for fan-out/independent parallel REFACTOR branches
- team-coordinate: fix dependency_graph schema mismatch, needs_research sequencing,
  handleSpawnNext role→task level check, add output_tag to template, precise inner_loop rules

All handleSpawnNext now read task description InnerLoop: field instead of role-level default,
enabling same-role workers to run in parallel when tasks have no mutual dependencies.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

Author: catlog22

.claude/skills/team-arch-opt/SKILL.md

@@ -78,7 +78,7 @@ Execute built-in Phase 1 (task discovery) -> role Phase 2-4 -> built-in Phase 5
 })
 ```
 
-**Inner Loop roles** (refactorer): Set `inner_loop: true`.
+**Inner Loop roles** (refactorer): Set `inner_loop` dynamically — `true` for single mode, `false` for fan-out/independent (parallel branches).
 **Single-task roles** (analyzer, designer, validator, reviewer): Set `inner_loop: false`.
 
 ## User Commands

.claude/skills/team-arch-opt/roles/coordinator/commands/monitor.md

@@ -82,7 +82,9 @@ Find ready tasks, spawn workers, STOP.
 2. No ready + work in progress -> report waiting, STOP
 3. No ready + nothing in progress -> handleComplete
 4. Has ready -> for each:
-   a. Check if inner loop role with active worker -> skip (worker picks up)
+   a. Check inner_loop: parse task description `InnerLoop:` field (NOT role.md default)
+      - InnerLoop: true AND same-role worker already active -> skip (worker picks up)
+      - InnerLoop: false OR no active same-role worker -> spawn new worker
    b. TaskUpdate -> in_progress
    c. team_msg log -> task_unblocked
    d. Spawn team-worker (see SKILL.md Spawn Template):

.claude/skills/team-arch-opt/roles/refactorer/role.md

@@ -1,12 +1,14 @@
 ---
 role: refactorer
 prefix: REFACTOR
-inner_loop: true
+inner_loop: dynamic
 message_types: [state_update]
 ---
 
 # Code Refactorer
 
+> **inner_loop: dynamic** — Dispatch sets per-task: `true` for single mode (one REFACTOR task with iterative fix cycles), `false` for fan-out/independent modes (REFACTOR-B01..N run as separate parallel workers). When false, each branch gets its own worker.
+
 Implement architecture refactoring changes following the design plan. For FIX tasks, apply targeted corrections based on review/validation feedback.
 
 ## Modes

.claude/skills/team-arch-opt/specs/team-config.json

@@ -64,10 +64,10 @@
       "type": "code_generation",
       "description": "Implements architecture refactoring changes following the design plan",
       "role_spec": "roles/refactorer/role.md",
-      "inner_loop": true,
+      "inner_loop": "dynamic",
       "frontmatter": {
         "prefix": "REFACTOR",
-        "inner_loop": true,
+        "inner_loop": "dynamic",
         "additional_prefixes": ["FIX"],
         "discuss_rounds": [],
         "cli_tools": ["explore"],

.claude/skills/team-coordinate/SKILL.md

@@ -130,9 +130,10 @@ Execute built-in Phase 1 (task discovery) -> role-spec Phase 2-4 -> built-in Pha
 })
 ```
 
-**Inner Loop roles** (role has 2+ serial same-prefix tasks): Set `inner_loop: true`. The team-worker agent handles the loop internally.
-
-**Single-task roles**: Set `inner_loop: false`.
+**Inner Loop**: Determined per-task from task description `InnerLoop:` field, not per-role:
+- Serial chain (2+ tasks, each blockedBy previous): `inner_loop: true` — single worker loops
+- Parallel tasks (no mutual blockedBy): `inner_loop: false` — separate workers per task
+- Single-task roles: `inner_loop: false`
 
 ---
 

.claude/skills/team-coordinate/roles/coordinator/commands/analyze-task.md

@@ -130,7 +130,7 @@ For each role, determine frontmatter and generation hints:
 | Field | Derivation |
 |-------|------------|
 | `prefix` | From capability prefix (e.g., RESEARCH, DRAFT, IMPL) |
-| `inner_loop` | `true` if role has 2+ serial same-prefix tasks |
+| `inner_loop` | `true` if role has 2+ same-prefix tasks AND they form a serial chain (each blockedBy the previous). `false` if tasks are parallel (no mutual blockedBy) or role has only 1 task |
 | `CLI tools` | Suggested, not mandatory — coordinator may adjust based on task needs |
 | `pattern_hint` | Reference pattern name from role-spec-template (research/document/code/analysis/validation) — guides coordinator's Phase 2-4 composition, NOT a rigid template selector |
 | `output_type` | `artifact` (new files in session/artifacts/) / `codebase` (modify existing project files) / `mixed` (both) — determines verification strategy in Behavioral Traits |

.claude/skills/team-coordinate/roles/coordinator/commands/dispatch.md

@@ -95,9 +95,10 @@ RoleSpec: <session-folder>/role-specs/<role-name>.md
 
 | Condition | InnerLoop |
 |-----------|-----------|
-| Role has 2+ serial same-prefix tasks | true |
+| Role has 2+ same-prefix tasks forming a serial chain (each blockedBy the previous) | true |
 | Role has 1 task | false |
-| Tasks are parallel (no dependency between them) | false |
+| Role has 2+ same-prefix tasks but they are parallel (no mutual blockedBy) | false |
+| Mixed: some serial, some parallel within same role | Set per-task: true for serial chain members, false for parallel members |
 
 ### Dependency Validation
 

.claude/skills/team-coordinate/roles/coordinator/commands/monitor.md

@@ -144,9 +144,10 @@ Ready tasks found?
   +- NONE + work in progress -> report waiting -> STOP
   +- NONE + nothing in progress -> PIPELINE_COMPLETE -> handleComplete
   +- HAS ready tasks -> for each:
-      +- Is task owner an Inner Loop role AND that role already has an active_worker?
+      +- Parse task description `InnerLoop:` field (NOT session.roles[].inner_loop)
+      +- InnerLoop: true AND same-role worker already in active_workers?
       |   +- YES -> SKIP spawn (existing worker will pick it up via inner loop)
-      |   +- NO -> normal spawn below
+      |   +- NO -> normal spawn below (InnerLoop: false OR no active same-role worker)
       +- TaskUpdate -> in_progress
       +- team_msg log -> task_unblocked (session_id=<session-id>)
       +- Spawn team-worker (see spawn tool call below)

.claude/skills/team-coordinate/roles/coordinator/role.md

@@ -207,15 +207,9 @@ Regardless of complexity score or role count, coordinator MUST:
    - `project_root` = result of `Bash({ command: "pwd" })`
    - `skill_root` = `<project_root>/.claude/skills/team-coordinate`
 
-2. **Check `needs_research` flag** from task-analysis.json:
-   - If `true`: **Spawn researcher worker first** to gather codebase context
-     - Wait for researcher callback
-     - Merge research findings into task context
-     - Update task-analysis.json with enriched context
+2. **Generate session ID**: `TC-<slug>-<date>` (slug from first 3 meaningful words of task)
 
-3. **Generate session ID**: `TC-<slug>-<date>` (slug from first 3 meaningful words of task)
-
-4. **Create session folder structure**:
+3. **Create session folder structure**:
    ```
    .workflow/.team/<session-id>/
    +-- role-specs/
@@ -226,11 +220,11 @@ Regardless of complexity score or role count, coordinator MUST:
    +-- .msg/
    ```
 
-5. **Call TeamCreate** with team name derived from session ID
+4. **Call TeamCreate** with team name derived from session ID
 
-6. **Read `specs/role-spec-template.md`** for Behavioral Traits + Reference Patterns
+5. **Read `specs/role-spec-template.md`** for Behavioral Traits + Reference Patterns
 
-7. **For each role in task-analysis.json#roles**:
+6. **For each role in task-analysis.json#roles**:
    - Fill YAML frontmatter: role, prefix, inner_loop, additional_members, message_types
    - **Compose Phase 2-4 content** (NOT copy from template):
      - Phase 2: Derive input sources and context loading steps from **task description + upstream dependencies**
@@ -239,14 +233,14 @@ Regardless of complexity score or role count, coordinator MUST:
      - Reference Patterns may guide phase structure, but task description determines specific content
    - Write generated role-spec to `<session>/role-specs/<role-name>.md`
 
-8. **Register roles** in team-session.json#roles (with `role_spec` path instead of `role_file`)
+7. **Register roles** in team-session.json#roles (with `role_spec` path instead of `role_file`)
 
-9. **Initialize shared infrastructure**:
+8. **Initialize shared infrastructure**:
    - `wisdom/learnings.md`, `wisdom/decisions.md`, `wisdom/issues.md` (empty with headers)
    - `explorations/cache-index.json` (`{ "entries": [] }`)
    - `discussions/` (empty directory)
 
-10. **Initialize pipeline metadata** via team_msg:
+9. **Initialize pipeline metadata** via team_msg:
 ```typescript
 // 使用 team_msg 将 pipeline 元数据写入 .msg/meta.json
 // 注意: 此处为动态角色，执行时需将 <placeholders> 替换为 task-analysis.json 中生成的实际角色列表
@@ -265,7 +259,14 @@ mcp__ccw-tools__team_msg({
 })
 ```
 
-11. **Write team-session.json** with: session_id, task_description, status="active", roles, pipeline (empty), active_workers=[], completion_action="interactive", created_at
+10. **Write team-session.json** with: session_id, task_description, status="active", roles, pipeline (empty), active_workers=[], completion_action="interactive", created_at
+
+11. **Check `needs_research` flag** from task-analysis.json:
+    - If `true`: Spawn researcher worker (role-spec now exists from step 6) to gather codebase context
+      - Wait for researcher callback
+      - Merge research findings into task context
+      - Update task-analysis.json with enriched context
+    - If `false`: Skip, proceed to Phase 3
 
 **Success**: Session created, role-spec files generated, shared infrastructure initialized.
 

.claude/skills/team-coordinate/specs/pipelines.md

@@ -38,18 +38,20 @@ Phase 3-N: monitor.md
 
 ## Dependency Graph Structure
 
-task-analysis.json encodes the pipeline:
+task-analysis.json encodes the pipeline as adjacency list (task ID -> blockedBy array):
 
 ```json
 {
   "dependency_graph": {
-    "RESEARCH-001": { "role": "researcher", "blockedBy": [], "priority": "P0" },
-    "IMPL-001":     { "role": "developer",  "blockedBy": ["RESEARCH-001"], "priority": "P1" },
-    "TEST-001":     { "role": "tester",     "blockedBy": ["IMPL-001"], "priority": "P2" }
+    "RESEARCH-001": [],
+    "IMPL-001": ["RESEARCH-001"],
+    "TEST-001": ["IMPL-001"]
   }
 }
 ```
 
+Role mapping comes from `task-analysis.json#capabilities[].tasks[]`, not from the dependency graph itself.
+
 ## Role-Worker Map
 
 Dynamic — loaded from session role-specs at runtime:
@@ -64,6 +66,7 @@ Role-spec files contain YAML frontmatter:
 role: <role-name>
 prefix: <PREFIX>
 inner_loop: <true|false>
+output_tag: "[<role-name>]"
 message_types:
   success: <type>
   error: error