claude-code-workflow 7.2.30 → 7.3.0

package/.claude/commands/workflow-skill.md

@@ -0,0 +1,130 @@
+---
+name: workflow-skill
+description: Direct workflow skill launcher - invoke any registered skill by name with arguments
+argument-hint: "<skill-name> [arguments]"
+allowed-tools: Skill(*), AskUserQuestion(*), Read(*), Glob(*)
+---
+
+# Workflow Skill Launcher
+
+Direct entry point to invoke any workflow skill by name. Bypasses CCW intent analysis and routing — execute exactly the skill the user specifies.
+
+## Usage
+
+```
+/workflow-skill <skill-name> [arguments]
+```
+
+## Execution
+
+1. **Parse** `$ARGUMENTS` → extract `skill_name` (first token) and `args` (remaining)
+2. **Validate** skill exists (check skill list below)
+3. **Invoke** `Skill(skill_name, args)`
+
+```javascript
+const [skillName, ...rest] = $ARGUMENTS.trim().split(/\s+/);
+const args = rest.join(' ');
+
+// Validate
+if (!skillName) {
+  // Show skill catalog (see below)
+  return;
+}
+
+// Execute
+Skill(skillName, args);
+```
+
+## Skill Catalog
+
+If no arguments provided, show this catalog for user selection.
+
+### Planning Skills
+| Skill | Description |
+|-------|-------------|
+| `workflow-lite-plan` | Lightweight planning — explore → plan → confirm → handoff |
+| `workflow-plan` | Full planning — session → context → convention → gen → verify |
+| `workflow-tdd-plan` | TDD planning — 6-phase TDD plan → verify |
+| `workflow-multi-cli-plan` | Multi-CLI collaborative planning with codebase context |
+
+### Execution Skills
+| Skill | Description |
+|-------|-------------|
+| `workflow-lite-execute` | Lightweight execution — task grouping → batch → review → sync |
+| `workflow-execute` | Full execution — session discovery → task processing → commit |
+| `team-executor` | Resume existing team session and execute remaining tasks |
+| `team-planex` | Plan-and-execute wave pipeline (planner + executor) |
+
+### Review & Test Skills
+| Skill | Description |
+|-------|-------------|
+| `workflow-test-fix` | Test generation + fix cycle |
+| `workflow-lite-test-review` | Post-execution test review and fix |
+| `review-cycle` | Multi-dimensional code review with automated fix |
+| `review-code` | Code review with structured report (read-only) |
+
+### Exploration Skills
+| Skill | Description |
+|-------|-------------|
+| `brainstorm` | Multi-perspective brainstorming |
+| `spec-generator` | Specification generator — 7-phase document chain |
+| `investigate` | Systematic debugging with Iron Law methodology |
+
+### Workflow Commands (slash-style)
+| Skill | Description |
+|-------|-------------|
+| `workflow:analyze-with-file` | Collaborative analysis with documented discussions |
+| `workflow:brainstorm-with-file` | Interactive brainstorming with multi-CLI collaboration |
+| `workflow:collaborative-plan-with-file` | Multi-agent collaborative planning |
+| `workflow:debug-with-file` | Hypothesis-driven debugging |
+| `workflow:roadmap-with-file` | Strategic requirement roadmap |
+| `workflow:unified-execute-with-file` | Universal execution engine for any planning output |
+| `workflow:integration-test-cycle` | Self-iterating integration test workflow |
+| `workflow:refactor-cycle` | Tech debt discovery and refactoring cycle |
+| `workflow:clean` | Intelligent code cleanup |
+
+### Team Skills
+| Skill | Description |
+|-------|-------------|
+| `team-frontend` | Frontend development team |
+| `team-issue` | Issue resolution team |
+| `team-review` | Code review team (3-role pipeline) |
+| `team-testing` | Testing team (progressive coverage) |
+| `team-lifecycle-v4` | Full lifecycle team (plan → develop → test → review) |
+| `team-brainstorm` | Brainstorming team |
+| `team-arch-opt` | Architecture optimization team |
+| `team-perf-opt` | Performance optimization team |
+| `team-quality-assurance` | Quality assurance team |
+| `team-tech-debt` | Tech debt identification and removal |
+| `team-roadmap-dev` | Roadmap-driven development |
+| `team-uidesign` | UI design team |
+| `team-ui-polish` | UI polish team |
+| `team-ultra-analyze` | Deep collaborative analysis team |
+| `team-coordinate` | Universal team coordination with dynamic roles |
+
+### Meta Skills
+| Skill | Description |
+|-------|-------------|
+| `skill-generator` | Create new Claude Code skills |
+| `workflow-skill-designer` | Design orchestrator+phases structured skills |
+| `wf-composer` | Semantic workflow composer from natural language |
+| `wf-player` | Workflow template player |
+
+## Examples
+
+```bash
+# Direct skill invocation
+/workflow-skill workflow-lite-plan "Add user authentication"
+/workflow-skill workflow-plan "Refactor payment module"
+/workflow-skill brainstorm "Notification system architecture"
+/workflow-skill workflow-test-fix --session WFS-123
+/workflow-skill team-planex "Implement OAuth2 + 2FA"
+/workflow-skill workflow:analyze-with-file "认证架构分析"
+/workflow-skill review-cycle
+
+# With auto mode
+/workflow-skill workflow-lite-plan -y "Fix login timeout"
+
+# No arguments → show catalog
+/workflow-skill
+```

package/.claude/skills/ccw-chain/SKILL.md

@@ -6,69 +6,31 @@ allowed-tools: Skill(*), TodoWrite(*), AskUserQuestion(*), Read(*), Grep(*), Glo
 
 # CCW Chain Orchestrator
 
-Chain-based version of the CCW workflow orchestrator. Uses the chain_loader tool for progressive step loading and LLM-driven decision routing. Each step node loads the **actual command/skill document** for that pipeline stage.
+Chain-based workflow orchestrator using `chain_loader` for progressive step loading and LLM-driven decision routing.
 
-## Architecture
+## Discovery
 
-```
-User Input → ccw-main (intent → clarity → route)
-                        ↓
-         ccw-{category} (select flow → step sequence)
-                        ↓
-         Each step loads actual skill/command doc from phases/
-                        ↓
-         LLM executes via Skill() call per loaded doc
-```
-
-## Chain Files (8 chains)
-
-| Chain | Nodes | Purpose |
-|-------|-------|---------|
-| `ccw-main` | 5 | Entry: intent analysis → clarity check → category routing |
-| `ccw-lightweight` | 8 | Level 2: rapid, bugfix, hotfix, docs |
-| `ccw-standard` | 17 | Level 3: coupled, tdd, test-fix, review, multi-cli, ui |
-| `ccw-exploration` | 22 | Level 4: brainstorm, spec-driven, full, greenfield |
-| `ccw-with-file` | 9 | With-File: analyze, debug, collab-plan, roadmap |
-| `ccw-cycle` | 4 | Cycle: integration-test, refactor |
-| `ccw-issue` | 13 | Issue: batch, rapid-to-issue, brainstorm-to-issue |
-| `ccw-team` | 1 | Team: team-planex |
-
-## phases/ — Actual Command/Skill Documents
-
-Step nodes reference actual docs via `content_ref`. The `phases/` directory contains:
-
-**Commands** (flat files — copied from `.claude/commands/`):
-- `analyze-with-file.md`, `brainstorm-with-file.md`, `collaborative-plan-with-file.md`
-- `debug-with-file.md`, `integration-test-cycle.md`, `refactor-cycle.md`
-- `roadmap-with-file.md`, `unified-execute-with-file.md`, `ui-design-explore-auto.md`
-- `issue-discover.md`, `issue-plan.md`, `issue-queue.md`, `issue-execute.md`
-- `issue-convert-to-plan.md`, `issue-from-brainstorm.md`
-
-**Skills without phases** (flat files — copied from SKILL.md):
-- `workflow-lite-plan.md`, `workflow-multi-cli-plan.md`, `team-planex.md`
-
-**Skills with phases** (subdirectories — SKILL.md + phases/):
-- `workflow-plan/`, `workflow-execute/`, `workflow-tdd-plan/`
-- `workflow-test-fix/`, `review-cycle/`, `brainstorm/`, `spec-generator/`
+1. `chain_loader list` — list all chains with triggers, entries, and descriptions
+2. Match user intent to chain `triggers.task_types` / `triggers.keywords`
+3. `chain_loader inspect` — preview chain node graph and available entries
+4. `chain_loader start` — begin from default entry, named entry (`entry_name`), or any node (`node`)
 
 ## Execution Protocol
 
-When chain_loader delivers a step node with a skill/command doc:
+When `chain_loader` delivers a step node with a skill/command doc:
 
 1. **Read** the loaded doc content to understand the skill's purpose and interface
 2. **Assemble** the Skill call: `Skill(skill_name, args)`
-   - First step in pipeline: `args = "${analysis.goal}"`
-   - Subsequent steps: `args = ""` (auto-receives session context from prior step)
+   - First step: `args = "${analysis.goal}"`
+   - Subsequent steps: `args = ""` (auto-receives session context)
    - Special args noted in step name (e.g., `--bugfix`, `--hotfix`, `--plan-only`)
 3. **Propagate -y**: If auto mode active, append `-y` to args
 4. **Execute**: `Skill(skill_name, args)` — blocking, wait for completion
-5. **Track**: Update TodoWrite status, proceed to next step via `done`
+5. **Advance**: `chain_loader done` to proceed to next step
 
 ```javascript
-// Auto mode detection
 const autoYes = /\b(-y|--yes)\b/.test($ARGUMENTS);
 
-// Skill call assembly with -y propagation
 function assembleCommand(skillName, args, previousResult) {
   if (!args && previousResult?.session_id) {
     args = `--session="${previousResult.session_id}"`;
@@ -87,33 +49,44 @@ function assembleCommand(skillName, args, previousResult) {
 - Error handling: auto-skip failed steps, continue pipeline
 - Propagation: `-y` injected into every downstream Skill call
 
-## Intent Analysis (ccw-main S1)
+## Delegation Protocol
 
-Detect task_type by regex priority — see `specs/intent-patterns.md` for full patterns.
+When `chain_loader` returns `delegate_depth > 0`:
+1. Continue normal execution (read content, assemble Skill, execute)
+2. On `returned_from_delegate: true`, resume parent chain context
+3. Variables received from child chain are available for subsequent steps
 
-Key mappings: task_type → category chain → specific flow
+## Preloaded Context
 
-## Execution via chain_loader
+When `chain_loader start` returns `preloaded_keys`:
+1. Preloaded content is available via `chain_loader content` for the entire session
+2. Reference preloaded context when assembling Skill calls
+3. Use preloaded memory/project context to inform all downstream steps
 
-```bash
-# Start
-ccw tool exec chain_loader '{"cmd":"start","skill":"ccw-chain","chain":"ccw-main"}'
+## Progress Visualization
 
-# Progress through steps
-ccw tool exec chain_loader '{"cmd":"done","session_id":"CL-ccw-chain-..."}'
+After each `chain_loader done`, call `chain_loader visualize` to show progress.
+Display the visualization in execution log for user awareness.
 
-# Decision nodes: pass choice index (1-based)
-ccw tool exec chain_loader '{"cmd":"done","session_id":"CL-ccw-chain-...","choice":1}'
-```
+## Variable Propagation
+
+Intent analysis results (`task_type`, `goal`, `auto_yes`) are stored as chain variables.
+`assembleCommand()` reads variables from `chain_loader status` for Skill args.
+Variables automatically flow through delegation via `pass_variables`/`receive_variables`.
+
+## Phase-Level Execution (Skill Chain Delegation)
+
+When the current chain is a skill-level chain (entered via delegation from a category chain):
+1. Each step delivers **phase doc content** directly (not SKILL.md)
+2. **Execute phase instructions inline** — do NOT wrap in `Skill()` call
+3. Reference preloaded `skill-context` for orchestration patterns (TodoWrite, data flow, error handling)
+4. Phase execution produces artifacts (files, session state) consumed by the next phase
+5. The chain system controls phase progression — no need for internal phase orchestration
+
+## Architecture: Chain Definition Layers
 
-## Pipeline Examples
-
-| Input | Chain Path | Pipeline |
-|-------|-----------|----------|
-| "Add API endpoint" | main→lightweight→Rapid | lite-plan → test-fix |
-| "Fix login timeout" | main→lightweight→Bugfix | lite-plan(--bugfix) → test-fix |
-| "OAuth2 system" (high) | main→standard→Coupled | plan → execute → review → test-fix |
-| "brainstorm: notifications" | main→exploration→Brainstorm | brainstorm-file → plan → execute → test-fix |
-| "debug WebSocket" | main→with-file→Debug | debug-with-file |
-| "integration test: payments" | main→cycle→Integration | integration-test-cycle |
-| "roadmap: OAuth + 2FA" | main→with-file→Roadmap | roadmap-file → team-planex |
+- **Category chains** (8): `ccw-chain/chains/` — routing and orchestration (ccw-main, ccw-standard, etc.)
+- **Workflow skill chains** (7): `.claude/workflow-skills/*/chains/` — skill-level chains with phase content
+- **Phase content**: `.claude/skills/*/phases/` — original phase files, referenced via `@skills/` prefix
+- Category chains delegate to workflow skill chains via `findChainAcrossSkills()` fallback
+- Content refs: `@phases/` = skill-relative, `@skills/` = project `.claude/skills/` relative

package/.claude/skills/ccw-chain/chains/ccw-cycle.json

@@ -2,8 +2,18 @@
   "chain_id": "ccw-cycle",
   "name": "CCW Cycle Workflows",
   "description": "Self-iterating cycle workflows: integration-test-cycle, refactor-cycle",
-  "version": "2.0",
+  "version": "2.1",
   "entry": "D1",
+  "entries": [
+    { "name": "default", "node": "D1", "description": "Select flow via decision node" },
+    { "name": "integration-test", "node": "S_IT1", "description": "Direct: Integration Test Cycle" },
+    { "name": "refactor", "node": "S_RC1", "description": "Direct: Refactor Cycle" }
+  ],
+  "triggers": {
+    "task_types": ["integration-test", "refactor"],
+    "keywords": ["integration.*test|e2e.*test", "refactor|tech.*debt"],
+    "scope": "Cycle: self-iterating integration-test and refactor workflows"
+  },
   "nodes": {
     "D1": {
       "type": "decision",
@@ -15,7 +25,7 @@
       ],
       "default": "S_IT1"
     },
-    "S_IT1": { "type": "step", "name": "Integration Test: integration-test-cycle", "content_ref": "@phases/integration-test-cycle.md", "next": null },
-    "S_RC1": { "type": "step", "name": "Refactor: refactor-cycle", "content_ref": "@phases/refactor-cycle.md", "next": null }
+    "S_IT1": { "type": "step", "name": "Integration Test: integration-test-cycle", "content_ref": "@commands/workflow/integration-test-cycle.md", "next": null },
+    "S_RC1": { "type": "step", "name": "Refactor: refactor-cycle", "content_ref": "@commands/workflow/refactor-cycle.md", "next": null }
   }
 }

package/.claude/skills/ccw-chain/chains/ccw-exploration.json

@@ -2,8 +2,19 @@
   "chain_id": "ccw-exploration",
   "name": "CCW Exploration Workflows (Level 4)",
   "description": "Level 4 workflow routing: brainstorm, spec-driven, full exploration, greenfield",
-  "version": "2.0",
+  "version": "2.1",
   "entry": "D1",
+  "entries": [
+    { "name": "default", "node": "D1", "description": "Select flow via decision node" },
+    { "name": "brainstorm", "node": "S_BP1", "description": "Direct: Brainstorm → Plan flow" },
+    { "name": "spec-driven", "node": "S_SD1", "description": "Direct: Spec-Driven flow" },
+    { "name": "full", "node": "S_FE1", "description": "Direct: Full Exploration flow" }
+  ],
+  "triggers": {
+    "task_types": ["brainstorm", "spec-driven", "greenfield", "exploration"],
+    "keywords": ["brainstorm|ideation|creative thinking", "spec.*gen|specification|PRD", "from scratch|greenfield|new project", "uncertain|explore|research|what if"],
+    "scope": "Level 4: brainstorm, spec-driven, greenfield, full exploration"
+  },
   "nodes": {
     "D1": {
       "type": "decision",
@@ -18,30 +29,30 @@
       ],
       "default": "S_BP1"
     },
-    "S_BP1": { "type": "step", "name": "Brainstorm: brainstorm-with-file", "content_ref": "@phases/brainstorm-with-file.md", "next": "S_BP2" },
-    "S_BP2": { "type": "step", "name": "Brainstorm: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_BP3" },
-    "S_BP3": { "type": "step", "name": "Brainstorm: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": "S_BP4" },
-    "S_BP4": { "type": "step", "name": "Brainstorm: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
+    "S_BP1": { "type": "step", "name": "Brainstorm: brainstorm-with-file", "content_ref": "@commands/workflow/brainstorm-with-file.md", "next": "S_BP2" },
+    "S_BP2": { "type": "delegate", "name": "Brainstorm: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_BP3" },
+    "S_BP3": { "type": "step", "name": "Brainstorm: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": "S_BP4" },
+    "S_BP4": { "type": "delegate", "name": "Brainstorm: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
 
-    "S_SD1": { "type": "step", "name": "Spec: spec-generator", "content_ref": "@phases/spec-generator/SKILL.md", "next": "S_SD2" },
-    "S_SD2": { "type": "step", "name": "Spec: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_SD3" },
-    "S_SD3": { "type": "step", "name": "Spec: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": "S_SD4" },
-    "S_SD4": { "type": "step", "name": "Spec: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
+    "S_SD1": { "type": "delegate", "name": "Spec: spec-generator", "chain": "spec-generator", "pass_variables": ["auto_yes", "goal"], "next": "S_SD2" },
+    "S_SD2": { "type": "delegate", "name": "Spec: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_SD3" },
+    "S_SD3": { "type": "step", "name": "Spec: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": "S_SD4" },
+    "S_SD4": { "type": "delegate", "name": "Spec: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
 
-    "S_FE1": { "type": "step", "name": "Full: brainstorm", "content_ref": "@phases/brainstorm/SKILL.md", "next": "S_FE2" },
-    "S_FE2": { "type": "step", "name": "Full: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_FE3" },
-    "S_FE3": { "type": "step", "name": "Full: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": "S_FE4" },
-    "S_FE4": { "type": "step", "name": "Full: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
+    "S_FE1": { "type": "delegate", "name": "Full: brainstorm", "chain": "brainstorm", "pass_variables": ["auto_yes", "goal"], "next": "S_FE2" },
+    "S_FE2": { "type": "delegate", "name": "Full: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_FE3" },
+    "S_FE3": { "type": "step", "name": "Full: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": "S_FE4" },
+    "S_FE4": { "type": "delegate", "name": "Full: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
 
-    "S_GM1": { "type": "step", "name": "Greenfield-M: brainstorm-with-file", "content_ref": "@phases/brainstorm-with-file.md", "next": "S_GM2" },
-    "S_GM2": { "type": "step", "name": "Greenfield-M: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_GM3" },
-    "S_GM3": { "type": "step", "name": "Greenfield-M: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": "S_GM4" },
-    "S_GM4": { "type": "step", "name": "Greenfield-M: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
+    "S_GM1": { "type": "step", "name": "Greenfield-M: brainstorm-with-file", "content_ref": "@commands/workflow/brainstorm-with-file.md", "next": "S_GM2" },
+    "S_GM2": { "type": "delegate", "name": "Greenfield-M: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_GM3" },
+    "S_GM3": { "type": "step", "name": "Greenfield-M: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": "S_GM4" },
+    "S_GM4": { "type": "delegate", "name": "Greenfield-M: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
 
-    "S_GL1": { "type": "step", "name": "Greenfield-L: brainstorm-with-file", "content_ref": "@phases/brainstorm-with-file.md", "next": "S_GL2" },
-    "S_GL2": { "type": "step", "name": "Greenfield-L: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_GL3" },
-    "S_GL3": { "type": "step", "name": "Greenfield-L: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": "S_GL4" },
-    "S_GL4": { "type": "step", "name": "Greenfield-L: review-cycle", "content_ref": "@phases/review-cycle/SKILL.md", "next": "S_GL5" },
-    "S_GL5": { "type": "step", "name": "Greenfield-L: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null }
+    "S_GL1": { "type": "step", "name": "Greenfield-L: brainstorm-with-file", "content_ref": "@commands/workflow/brainstorm-with-file.md", "next": "S_GL2" },
+    "S_GL2": { "type": "delegate", "name": "Greenfield-L: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_GL3" },
+    "S_GL3": { "type": "step", "name": "Greenfield-L: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": "S_GL4" },
+    "S_GL4": { "type": "delegate", "name": "Greenfield-L: review-cycle", "chain": "ccw-review-cycle", "pass_variables": ["auto_yes", "goal"], "next": "S_GL5" },
+    "S_GL5": { "type": "delegate", "name": "Greenfield-L: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null }
   }
 }

package/.claude/skills/ccw-chain/chains/ccw-issue.json

@@ -2,8 +2,19 @@
   "chain_id": "ccw-issue",
   "name": "CCW Issue Workflows",
   "description": "Issue management workflows: batch processing, rapid-to-issue bridge, brainstorm-to-issue conversion",
-  "version": "2.0",
+  "version": "2.1",
   "entry": "D1",
+  "entries": [
+    { "name": "default", "node": "D1", "description": "Select flow via decision node" },
+    { "name": "batch", "node": "S_IB1", "description": "Direct: Issue Batch (discover → plan → queue → execute)" },
+    { "name": "rapid-to-issue", "node": "S_RI1", "description": "Direct: Rapid-to-Issue bridge" },
+    { "name": "brainstorm-to-issue", "node": "S_BI1", "description": "Direct: Brainstorm-to-Issue" }
+  ],
+  "triggers": {
+    "task_types": ["issue-batch", "issue-transition", "brainstorm-to-issue"],
+    "keywords": ["issues?.*batch|batch.*fix|batch.*resolve", "issue workflow|structured workflow|queue", "brainstorm.*issue|idea.*issue"],
+    "scope": "Issue: batch processing, rapid-to-issue, brainstorm-to-issue conversion"
+  },
   "nodes": {
     "D1": {
       "type": "decision",
@@ -16,18 +27,18 @@
       ],
       "default": "S_IB1"
     },
-    "S_IB1": { "type": "step", "name": "Issue Batch: issue-discover", "content_ref": "@phases/issue-discover.md", "next": "S_IB2" },
-    "S_IB2": { "type": "step", "name": "Issue Batch: issue-plan", "content_ref": "@phases/issue-plan.md", "next": "S_IB3" },
-    "S_IB3": { "type": "step", "name": "Issue Batch: issue-queue", "content_ref": "@phases/issue-queue.md", "next": "S_IB4" },
-    "S_IB4": { "type": "step", "name": "Issue Batch: issue-execute", "content_ref": "@phases/issue-execute.md", "next": null },
+    "S_IB1": { "type": "step", "name": "Issue Batch: issue-discover", "content_ref": "@commands/issue/discover.md", "next": "S_IB2" },
+    "S_IB2": { "type": "step", "name": "Issue Batch: issue-plan", "content_ref": "@commands/issue/plan.md", "next": "S_IB3" },
+    "S_IB3": { "type": "step", "name": "Issue Batch: issue-queue", "content_ref": "@commands/issue/queue.md", "next": "S_IB4" },
+    "S_IB4": { "type": "step", "name": "Issue Batch: issue-execute", "content_ref": "@commands/issue/execute.md", "next": null },
 
-    "S_RI1": { "type": "step", "name": "Rapid-to-Issue: workflow-lite-plan", "content_ref": "@phases/workflow-lite-plan.md", "next": "S_RI2" },
-    "S_RI2": { "type": "step", "name": "Rapid-to-Issue: issue-convert-to-plan", "content_ref": "@phases/issue-convert-to-plan.md", "next": "S_RI3" },
-    "S_RI3": { "type": "step", "name": "Rapid-to-Issue: issue-queue", "content_ref": "@phases/issue-queue.md", "next": "S_RI4" },
-    "S_RI4": { "type": "step", "name": "Rapid-to-Issue: issue-execute", "content_ref": "@phases/issue-execute.md", "next": null },
+    "S_RI1": { "type": "step", "name": "Rapid-to-Issue: workflow-lite-plan", "content_ref": "@skills/workflow-lite-plan/SKILL.md", "next": "S_RI2" },
+    "S_RI2": { "type": "step", "name": "Rapid-to-Issue: issue-convert-to-plan", "content_ref": "@commands/issue/convert-to-plan.md", "next": "S_RI3" },
+    "S_RI3": { "type": "step", "name": "Rapid-to-Issue: issue-queue", "content_ref": "@commands/issue/queue.md", "next": "S_RI4" },
+    "S_RI4": { "type": "step", "name": "Rapid-to-Issue: issue-execute", "content_ref": "@commands/issue/execute.md", "next": null },
 
-    "S_BI1": { "type": "step", "name": "Brainstorm-to-Issue: from-brainstorm", "content_ref": "@phases/issue-from-brainstorm.md", "next": "S_BI2" },
-    "S_BI2": { "type": "step", "name": "Brainstorm-to-Issue: issue-queue", "content_ref": "@phases/issue-queue.md", "next": "S_BI3" },
-    "S_BI3": { "type": "step", "name": "Brainstorm-to-Issue: issue-execute", "content_ref": "@phases/issue-execute.md", "next": null }
+    "S_BI1": { "type": "step", "name": "Brainstorm-to-Issue: from-brainstorm", "content_ref": "@commands/issue/from-brainstorm.md", "next": "S_BI2" },
+    "S_BI2": { "type": "step", "name": "Brainstorm-to-Issue: issue-queue", "content_ref": "@commands/issue/queue.md", "next": "S_BI3" },
+    "S_BI3": { "type": "step", "name": "Brainstorm-to-Issue: issue-execute", "content_ref": "@commands/issue/execute.md", "next": null }
   }
 }

package/.claude/skills/ccw-chain/chains/ccw-lightweight.json

@@ -2,8 +2,20 @@
   "chain_id": "ccw-lightweight",
   "name": "CCW Lightweight Workflows (Level 2)",
   "description": "Level 2 workflow routing: rapid, bugfix, hotfix, docs — simple single/dual-skill pipelines",
-  "version": "2.0",
+  "version": "2.1",
   "entry": "D1",
+  "entries": [
+    { "name": "default", "node": "D1", "description": "Select flow via decision node" },
+    { "name": "rapid", "node": "S_R1", "description": "Direct: Rapid flow (lite-plan → test-fix)" },
+    { "name": "bugfix", "node": "S_B1", "description": "Direct: Bugfix flow (lite-plan --bugfix → test-fix)" },
+    { "name": "hotfix", "node": "S_H1", "description": "Direct: Hotfix flow (lite-plan --hotfix)" },
+    { "name": "docs", "node": "S_D1", "description": "Direct: Docs flow (lite-plan)" }
+  ],
+  "triggers": {
+    "task_types": ["bugfix", "bugfix-hotfix", "quick-task", "documentation", "feature"],
+    "keywords": ["fix|bug|error|crash|fail", "quick|simple|small", "docs|documentation|readme", "urgent|production|critical"],
+    "scope": "Level 2: rapid, bugfix, hotfix, docs — simple single/dual-skill pipelines"
+  },
   "nodes": {
     "D1": {
       "type": "decision",
@@ -20,37 +32,39 @@
     "S_R1": {
       "type": "step",
       "name": "Rapid: workflow-lite-plan",
-      "content_ref": "@phases/workflow-lite-plan.md",
+      "content_ref": "@skills/workflow-lite-plan/SKILL.md",
       "next": "S_R2"
     },
     "S_R2": {
-      "type": "step",
+      "type": "delegate",
       "name": "Rapid: workflow-test-fix",
-      "content_ref": "@phases/workflow-test-fix/SKILL.md",
+      "chain": "workflow-test-fix",
+      "pass_variables": ["auto_yes", "goal"],
       "next": null
     },
     "S_B1": {
       "type": "step",
       "name": "Bugfix: workflow-lite-plan --bugfix",
-      "content_ref": "@phases/workflow-lite-plan.md",
+      "content_ref": "@skills/workflow-lite-plan/SKILL.md",
       "next": "S_B2"
     },
     "S_B2": {
-      "type": "step",
+      "type": "delegate",
       "name": "Bugfix: workflow-test-fix",
-      "content_ref": "@phases/workflow-test-fix/SKILL.md",
+      "chain": "workflow-test-fix",
+      "pass_variables": ["auto_yes", "goal"],
       "next": null
     },
     "S_H1": {
       "type": "step",
       "name": "Hotfix: workflow-lite-plan --hotfix",
-      "content_ref": "@phases/workflow-lite-plan.md",
+      "content_ref": "@skills/workflow-lite-plan/SKILL.md",
       "next": null
     },
     "S_D1": {
       "type": "step",
       "name": "Docs: workflow-lite-plan",
-      "content_ref": "@phases/workflow-lite-plan.md",
+      "content_ref": "@skills/workflow-lite-plan/SKILL.md",
       "next": null
     }
   }

package/.claude/skills/ccw-chain/chains/ccw-main.json

@@ -2,13 +2,26 @@
   "chain_id": "ccw-main",
   "name": "CCW Main Orchestrator",
   "description": "Main workflow orchestrator: intent analysis → clarity check → workflow category routing",
-  "version": "2.0",
+  "version": "2.2",
+  "preload": [
+    { "key": "memory", "source": "memory:MEMORY.md", "required": false },
+    { "key": "chain-list", "source": "@chains/_index.md", "required": false }
+  ],
+  "variables": {
+    "task_type": { "type": "string", "description": "Detected task type" },
+    "goal": { "type": "string", "description": "User's goal" },
+    "auto_yes": { "type": "boolean", "default": false, "description": "Auto-confirm mode" }
+  },
   "entry": "S1",
+  "entries": [
+    { "name": "default", "node": "S1", "description": "Full flow: intent analysis → clarity check → routing" },
+    { "name": "skip-analysis", "node": "D2", "description": "Skip analysis, directly select workflow category" }
+  ],
   "nodes": {
     "S1": {
       "type": "step",
       "name": "Analyze Intent",
-      "content_inline": "## Phase 1: Analyze Intent\n\nAnalyze the user's input ($ARGUMENTS) to determine:\n\n```javascript\nconst analysis = {\n  goal: extractGoal(input),\n  scope: extractScope(input),\n  constraints: extractConstraints(input),\n  task_type: detectTaskType(input),\n  complexity: assessComplexity(input),  // low|medium|high\n  clarity_score: calculateClarity(input) // 0-3 (>=2 = clear)\n};\n\nconst autoYes = /\\b(-y|--yes)\\b/.test($ARGUMENTS);\n```\n\nDetect task_type by priority (see specs/intent-patterns.md for full regex list):\n- bugfix-hotfix, greenfield, brainstorm, brainstorm-to-issue, debug-file, analyze-file\n- collaborative-plan, roadmap, spec-driven, integration-test, refactor, team-planex\n- multi-cli, bugfix, issue-batch, issue-transition, exploration, quick-task\n- ui-design, tdd, test-fix, test-gen, review, documentation, feature (default)\n\nOutput: `Type: [task_type] | Goal: [goal] | Complexity: [complexity] | Clarity: [clarity_score]/3`",
+      "content_inline": "## Phase 1: Analyze Intent\n\nAnalyze the user's input ($ARGUMENTS) to determine:\n\n```javascript\nconst analysis = {\n  goal: extractGoal(input),\n  scope: extractScope(input),\n  constraints: extractConstraints(input),\n  task_type: detectTaskType(input),\n  complexity: assessComplexity(input),  // low|medium|high\n  clarity_score: calculateClarity(input) // 0-3 (>=2 = clear)\n};\n\nconst autoYes = /\\b(-y|--yes)\\b/.test($ARGUMENTS);\n```\n\nDetect task_type by priority (use `chain_loader list` to see each chain's triggers):\n- bugfix-hotfix, greenfield, brainstorm, brainstorm-to-issue, debug-file, analyze-file\n- collaborative-plan, roadmap, spec-driven, integration-test, refactor, team-planex\n- multi-cli, bugfix, issue-batch, issue-transition, exploration, quick-task\n- ui-design, tdd, test-fix, test-gen, review, documentation, feature (default)\n\nOutput: `Type: [task_type] | Goal: [goal] | Complexity: [complexity] | Clarity: [clarity_score]/3`",
       "next": "D1"
     },
     "D1": {

package/.claude/skills/ccw-chain/chains/ccw-standard.json

@@ -2,8 +2,20 @@
   "chain_id": "ccw-standard",
   "name": "CCW Standard Workflows (Level 3)",
   "description": "Level 3 workflow routing: coupled, tdd, test-fix, review, multi-cli, ui",
-  "version": "2.0",
+  "version": "2.2",
   "entry": "D1",
+  "entries": [
+    { "name": "default", "node": "D1", "description": "Select flow via decision node" },
+    { "name": "coupled", "node": "S_C1", "description": "Direct: Coupled feature (plan → execute → review → test-fix)" },
+    { "name": "tdd", "node": "S_T1", "description": "Direct: TDD (tdd-plan → execute)" },
+    { "name": "test-gen", "node": "S_TG1", "description": "Direct: Test generation" },
+    { "name": "review", "node": "S_RV1", "description": "Direct: Review cycle + fix" }
+  ],
+  "triggers": {
+    "task_types": ["tdd", "test-fix", "test-gen", "review", "multi-cli", "ui-design"],
+    "keywords": ["tdd|test-driven|test first", "test fail|fix test|failing test", "generate test|add test", "review|code review", "multi.*cli|multi.*model", "ui|design|component|style"],
+    "scope": "Level 3: coupled feature, TDD, test-fix, review, multi-cli, UI design"
+  },
   "nodes": {
     "D1": {
       "type": "decision",
@@ -20,20 +32,20 @@
       ],
       "default": "S_C1"
     },
-    "S_C1": { "type": "step", "name": "Coupled: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_C2" },
-    "S_C2": { "type": "step", "name": "Coupled: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": "S_C3" },
-    "S_C3": { "type": "step", "name": "Coupled: review-cycle", "content_ref": "@phases/review-cycle/SKILL.md", "next": "S_C4" },
-    "S_C4": { "type": "step", "name": "Coupled: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
-    "S_T1": { "type": "step", "name": "TDD: workflow-tdd-plan", "content_ref": "@phases/workflow-tdd-plan/SKILL.md", "next": "S_T2" },
-    "S_T2": { "type": "step", "name": "TDD: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": null },
-    "S_TG1": { "type": "step", "name": "Test Gen: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
-    "S_TF1": { "type": "step", "name": "Test Fix: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
-    "S_RF1": { "type": "step", "name": "Review: review-cycle", "content_ref": "@phases/review-cycle/SKILL.md", "next": "S_RF2" },
-    "S_RF2": { "type": "step", "name": "Review: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
-    "S_MC1": { "type": "step", "name": "Multi-CLI: workflow-multi-cli-plan", "content_ref": "@phases/workflow-multi-cli-plan.md", "next": "S_MC2" },
-    "S_MC2": { "type": "step", "name": "Multi-CLI: workflow-test-fix", "content_ref": "@phases/workflow-test-fix/SKILL.md", "next": null },
-    "S_UI1": { "type": "step", "name": "UI: ui-design-explore-auto", "content_ref": "@phases/ui-design-explore-auto.md", "next": "S_UI2" },
-    "S_UI2": { "type": "step", "name": "UI: workflow-plan", "content_ref": "@phases/workflow-plan/SKILL.md", "next": "S_UI3" },
-    "S_UI3": { "type": "step", "name": "UI: workflow-execute", "content_ref": "@phases/workflow-execute/SKILL.md", "next": null }
+    "S_C1": { "type": "delegate", "name": "Coupled: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "receive_variables": ["session_id"], "next": "S_C2" },
+    "S_C2": { "type": "step", "name": "Coupled: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": "S_C3" },
+    "S_C3": { "type": "delegate", "name": "Coupled: review-cycle", "chain": "ccw-review-cycle", "pass_variables": ["auto_yes", "goal"], "receive_variables": ["review_status", "fix_count"], "next": "S_C4" },
+    "S_C4": { "type": "delegate", "name": "Coupled: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal", "session_id"], "next": null },
+    "S_T1": { "type": "delegate", "name": "TDD: workflow-tdd-plan", "chain": "workflow-tdd-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_T2" },
+    "S_T2": { "type": "step", "name": "TDD: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": null },
+    "S_TG1": { "type": "delegate", "name": "Test Gen: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
+    "S_TF1": { "type": "delegate", "name": "Test Fix: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
+    "S_RF1": { "type": "delegate", "name": "Review: review-cycle", "chain": "ccw-review-cycle", "pass_variables": ["auto_yes", "goal"], "next": "S_RF2" },
+    "S_RF2": { "type": "delegate", "name": "Review: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
+    "S_MC1": { "type": "step", "name": "Multi-CLI: workflow-multi-cli-plan", "content_ref": "@skills/workflow-multi-cli-plan/SKILL.md", "next": "S_MC2" },
+    "S_MC2": { "type": "delegate", "name": "Multi-CLI: workflow-test-fix", "chain": "workflow-test-fix", "pass_variables": ["auto_yes", "goal"], "next": null },
+    "S_UI1": { "type": "step", "name": "UI: ui-design-explore-auto", "content_ref": "@commands/workflow/ui-design/explore-auto.md", "next": "S_UI2" },
+    "S_UI2": { "type": "delegate", "name": "UI: workflow-plan", "chain": "workflow-plan", "pass_variables": ["auto_yes", "goal"], "next": "S_UI3" },
+    "S_UI3": { "type": "step", "name": "UI: workflow-execute", "content_ref": "@skills/workflow-execute/SKILL.md", "next": null }
   }
 }