speccrew 0.7.74 → 0.7.76

package/.speccrew/skills/speccrew-pm-phase2-complexity-assess/SKILL.md

@@ -1,111 +1,15 @@
 ---
 name: speccrew-pm-phase2-complexity-assess
-description: Complexity assessment and skill routing for PM Phase 2. Validates knowledge initialization completeness, evaluates requirement complexity across 5 dimensions, and routes to appropriate PRD workflow (simple vs full pipeline). Use after Phase 1 knowledge detection completes.
+description: Complexity assessment and skill routing for PM Phase 2. Validates knowledge initialization completeness, evaluates requirement complexity across 5 dimensions, and routes to appropriate PRD workflow.
 tools: Read, Grep
 ---
 
-# Phase 2: Complexity Assessment & Skill Routing
-
-Assess requirement complexity and determine the appropriate skill routing path for PRD generation.
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - "Assess requirement complexity"
 - "Determine PRD workflow path"
-- "Should I use simple or complex PRD workflow?"
 - "Route to appropriate PRD skill"
-- "Check if knowledge initialization is complete"
-
-## User
-
-PM Agent (speccrew-product-manager)
-
-## Input
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| user_requirement | string | Yes | User requirement description or document path |
-| workspace_path | string | Yes | Absolute path to speccrew-workspace root |
-| knowledge_status | object | Yes | Knowledge status from Phase 1 (status, system_overview_path) |
-| language | string | No | Target language for output (auto-detected) |
-
-## Output
-
-| Field | Type | Description |
-|-------|------|-------------|
-| complexity_level | string | simple / moderate / complex |
-| route | string | speccrew-pm-requirement-simple or full-pipeline |
-| complexity_details | object | Detailed breakdown of 5-dimension assessment |
-| verification_status | string | Knowledge initialization verification result |
 
 ## AgentFlow Definition
-
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-## Complexity Assessment Matrix
-
-### 5-Dimension Evaluation
-
-| Dimension | Simple | Moderate | Complex |
-|-----------|--------|----------|---------|
-| **Functional Scope** | 1-2 modules, 1-5 features | 2-3 modules, 5-10 features | 3+ modules, 6+ features |
-| **Technical Complexity** | No new tech, 0-2 integrations | 1 new tech, 2-4 integrations | 2+ new techs, 5+ integrations |
-| **Business Rules** | Simple conditions, no state machine | Moderate branching, basic state machine | Complex branching, full state machine |
-| **Data Complexity** | 1-3 entities, simple relationships | 4-6 entities, moderate relationships | 7+ entities, complex relationships |
-| **Team/Time Constraints** | Standard timeline, single team | Moderate constraints | Tight deadline, multiple teams |
-
-### Decision Rules
-
-**→ Simple** (ANY of these):
-- Adding/modifying fields on an existing page
-- Minor feature enhancement within 1-2 modules
-- Business logic adjustment
-- Bug fix documentation
-- Scope: ≤ 5 features, ≤ 2 modules
-
-**→ Moderate** (ANY of these):
-- 2-3 modules involved
-- 5-10 features expected
-- Some cross-module dependencies
-- Standard new feature development
-
-**→ Complex** (ANY of these):
-- New system or major subsystem development
-- Involves 3+ modules
-- Requires 6+ features
-- Needs cross-module dependency management
-- User explicitly requests comprehensive analysis
-
-### Skill Routing
-
-| Complexity | Route | Workflow |
-|------------|-------|----------|
-| simple | speccrew-pm-requirement-simple | Single PRD, streamlined 6-step flow |
-| moderate | full-pipeline | clarify → model → analysis (Master-Sub PRD) |
-| complex | full-pipeline | clarify → model → analysis (Master-Sub PRD) |
-
-> **Default to Simple when in doubt**. It's easier to escalate from simple to complex than to simplify an over-engineered analysis.
-
-## Checklist
-
-- [ ] DISPATCH-PROGRESS.json exists and counts.pending == 0
-- [ ] Knowledge initialization verified complete
-- [ ] User requirement analyzed for module count
-- [ ] Functional scope assessed (modules, features)
-- [ ] Technical complexity evaluated (new tech, integrations)
-- [ ] Business rules complexity determined (conditions, state machine)
-- [ ] Data complexity assessed (entities, relationships)
-- [ ] Team/time constraints considered
-- [ ] Complexity level determined (simple/moderate/complex)
-- [ ] Skill route selected and output
-
-## Key Rules
-
-1. **MANDATORY Verification**: Phase 2.0 knowledge initialization verification CANNOT be skipped
-2. **Pending Tasks Block**: If `counts.pending > 0`, return to Phase 1 Path B - do NOT proceed
-3. **Read-Only Assessment**: This skill performs assessment only, no file modifications
-4. **Escalation Path**: Simple workflow can escalate to complex if scope expands during execution
-5. **Single Gateway**: Use exclusive gateway for routing decision - only one path selected
-6. **FORBIDDEN File Generation**: Phase 2 MUST NOT create any output files (evaluation reports, assessment JSON, markdown summaries, etc.). All output must be returned as variables only via the output block. Creating files outside of block definitions violates Strict Block Adherence.
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: speccrew-workspace/docs/rules/agentflow-spec.md
+> Then read and execute the XML workflow in SKILL.xml block-by-block as the authoritative execution plan.

package/.speccrew/skills/speccrew-pm-phase5-subprd-dispatch/SKILL.md

@@ -1,107 +1,15 @@
-# speccrew-pm-phase5-subprd-dispatch
-
-## Description
-
-PM Phase 5 Sub-PRD Batch Dispatch Skill. **Orchestration-layer skill** executed directly by PM Agent (NOT dispatched to Worker).
-
-This is an **orchestration skill** that PM Agent must execute directly:
-- PM Agent reads `SKILL.xml` and executes each block step-by-step
-- PM Agent uses Agent tool to dispatch Workers for Sub-PRD generation
-- Workers generate Sub-PRD files; PM Agent only coordinates
-
-**CRITICAL ARCHITECTURE RULE (Harness Principle 17: Orchestration Layer Separation)**:
-- This skill contains internal `dispatch-to-worker` blocks
-- Workers CANNOT dispatch Workers (execution hierarchy constraint)
-- Therefore, this skill MUST be executed by PM Agent directly
-- PM Agent MUST NOT dispatch this skill to a Worker
-
-**Execution Method**: PM Agent reads `SKILL.xml` and follows the workflow steps.
-
-## MANDATORY: PM Must Read SKILL.xml
-
-> 🛑 **BEFORE executing this skill, PM Agent MUST:**
-> 1. Read the `SKILL.xml` file in this skill directory
-> 2. Parse each block and execute in order
-> 3. Follow the exact workflow steps defined in the XML
->
-> **FORBIDDEN:**
-> - DO NOT execute this skill without reading the XML workflow
-> - DO NOT skip blocks defined in the XML
-> - DO NOT dispatch this skill to a Worker Agent
-
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `prd_output` | object | true | Phase 4 PRD generation output, containing master_prd_path, dispatch_plan_path, etc. |
-| `iteration_path` | string | true | Absolute path to current iteration directory |
-| `language` | string | false | User language (default: zh) |
-| `workspace_path` | string | true | Absolute path to speccrew-workspace root directory |
-| `update_progress_script` | string | true | Absolute path to update-progress.js script |
-| `max_concurrent_workers` | number | false | Maximum concurrent Workers (default: 5) |
-
-## Outputs
-
-| Output | Type | Description |
-|--------|------|-------------|
-| `dispatch_result` | string | Dispatch result status: success / partial / failed |
-| `total_subprds` | number | Total Sub-PRD count |
-| `success_count` | number | Successfully generated Sub-PRD count |
-| `failure_count` | number | Failed Sub-PRD count |
-| `sub_prd_files` | array | List of generated Sub-PRD file paths |
-| `feature_list_path` | string | Feature List file path |
-
+---
+name: speccrew-pm-phase5-subprd-dispatch
+description: PM Phase 5 Sub-PRD Batch Dispatch Skill. Orchestration-layer skill executed directly by PM Agent. Dispatches Workers for parallel Sub-PRD generation across modules.
+tools: Read, Write, Glob, Grep, Bash
+---
+
+# Trigger Scenarios
+- PM Agent executes Phase 5 Sub-PRD batch dispatch
+- Sub-PRD dispatch plan ready from Phase 4
+- Parallel Worker dispatch needed for multi-module PRD generation
+
+## AgentFlow Definition
 <!-- @agentflow: SKILL.xml -->
-
-## Checklist
-
-### Step 5.1: Read Dispatch Plan
-- [ ] Read dispatch-plan from Master PRD (Sub-PRD grouping information)
-- [ ] Parse sub_prd_groups array
-- [ ] Verify each group contains required fields: module_id, module_name, module_key, module_scope, module_entities
-
-### Step 5.2: Initialize Progress Tracking
-- [ ] Create temporary task definition file .tasks-temp.json
-- [ ] Initialize DISPATCH-PROGRESS.json using update-progress.js script
-- [ ] Verify initialization success (Total: N | Pending: N | Completed: 0)
-
-### Step 5.3: Batch Dispatch Workers
-- [ ] Group in batches of 5 for parallel dispatch
-- [ ] Each Worker carries complete context parameters
-- [ ] Update DISPATCH-PROGRESS.json after each batch completes
-- [ ] Continue to next batch until all modules are processed
-
-### Step 5.4: Failure Retry
-- [ ] Check for tasks with failed status
-- [ ] If failures exist -> retry once (single retry mechanism)
-- [ ] If still failing after retry -> log and continue
-
-### Step 5.5: Result Collection and Verification
-- [ ] Read final DISPATCH-PROGRESS.json
-- [ ] Verify all Sub-PRD files exist and size > 3KB
-- [ ] Generate summary report
-- [ ] Update checkpoint
-- [ ] Verify feature list completeness
-
-### Step 5.6: User Confirmation
-- [ ] Wait for user confirmation of Sub-PRD generation results
-
-## Must Do
-
-- **READ SKILL.xml FIRST** — PM Agent must parse the XML workflow before any execution
-- **Initialize DISPATCH-PROGRESS.json BEFORE any Worker dispatch** — Use update-progress.js init command
-- **Dispatch one Worker per module** — Use Agent tool to create speccrew-task-worker with speccrew-pm-sub-prd-generate skill
-- **Update progress after each Worker completes** — Use update-progress.js update-task command
-- **Verify all Sub-PRD files exist** — Check file existence and size > 3KB
-- **Update checkpoint via script** — Use update-progress.js write-checkpoint command
-
-## Must Not Do
-
-- **DO NOT dispatch this skill to a Worker** — Phase 5 is PM direct execution
-- **DO NOT skip reading SKILL.xml** — XML workflow is the execution guide
-- **DO NOT dispatch Workers before DISPATCH-PROGRESS.json is initialized** — Progress tracking is mandatory
-- **DO NOT create DISPATCH-PROGRESS.json manually** — Must use update-progress.js script
-- **DO NOT generate Sub-PRD content directly** — Only Workers generate Sub-PRDs
-- **DO NOT skip progress update after Worker completes** — Must call update-task for each completed Worker
-- **DO NOT dispatch one Worker for multiple modules** — One Worker per module
-- **DO NOT proceed to Phase 6 without verifying all Sub-PRDs** — Completion verification is mandatory
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: speccrew-workspace/docs/rules/agentflow-spec.md
+> Then read and execute the XML workflow in SKILL.xml block-by-block as the authoritative execution plan.

package/.speccrew/skills/speccrew-pm-phase6-verify-confirm/SKILL.md

@@ -1,92 +1,15 @@
 ---
 name: speccrew-pm-phase6-verify-confirm
-description: SpecCrew PM Phase 6 Verification & Confirmation Skill. Performs file integrity validation, feature list completeness check, user review gate, and finalization. Mandatory user confirmation required before status updates.
+description: SpecCrew PM Phase 6 Verification & Confirmation Skill. Performs file integrity validation, feature list completeness check, user review gate, and finalization.
 tools: Read, Write, Glob, Grep
 ---
 
-# Skill Overview
-
-Phase 6 Verification & Confirmation with three-stage gated execution. Ensures all PRD artifacts are valid, complete, and user-confirmed before finalizing workflow status.
-
-## Trigger Scenarios
-
-- PRD generation completed (Phase 4/5 done)
-- Sub-PRD dispatch completed (complex path)
-- Verification checklist needed before user review
+# Trigger Scenarios
+- PRD generation completed, verification needed
+- Sub-PRD dispatch completed, user review required
 - Final status update after user confirmation
 
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `iteration_path` | string | Yes | Path to iteration directory |
-| `complexity` | enum | Yes | `simple` or `complex` |
-| `prd_output` | object | Yes | PRD generation output containing file paths |
-| `dispatch_plan` | object | No | Dispatch plan for complex path (module count reference) |
-| `update_progress_script` | string | Yes | Path to update-progress.js script |
-| `language` | string | No | User language (auto-detected) |
-
-## Methodology Foundation
-
-Applies three-stage gated verification:
-- Stage 6.1: Automated file & content integrity validation
-- Stage 6.2: Mandatory user review and explicit confirmation
-- Stage 6.3: Finalization with status updates and cleanup
-
-## PM Stage Content Boundary
-
-> **DO NOT:** Update checkpoints before user confirmation, skip user review gate, auto-proceed without explicit "confirm" or "OK", modify PRD document status before confirmation.
-
-## Templates Used
-
-| Template | Path | Purpose |
-|----------|------|---------|
-| Verification Report | Inline | File integrity and feature completeness results |
-| PRD Review Summary | Inline | User review presentation template |
-
----
-
-# AgentFlow Definition
-
+## AgentFlow Definition
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
----
-
-# Output Checklist
-
-- [ ] Master PRD file verified (exists, size > 1KB)
-- [ ] All Sub-PRD files verified (complex path)
-- [ ] Sub-PRD count matches dispatch plan
-- [ ] `.clarification-summary.md` exists
-- [ ] `.module-design.md` exists (complex path)
-- [ ] Feature list completeness verified
-- [ ] User review gate executed
-- [ ] Explicit user confirmation received
-- [ ] Checkpoints updated via script
-- [ ] WORKFLOW-PROGRESS.json updated to completed
-- [ ] PRD document status updated to Confirmed
-- [ ] Intermediate files cleaned up
-
----
-
-# Constraints
-
-**Must do:**
-- Execute all verification steps in order: 6.1 → 6.2 → 6.3
-- Wait for explicit user confirmation ("confirm" or "OK") before Phase 6.3
-- Use `update-progress.js` for all JSON file updates
-- Present complete verification summary to user
-- Report all deliverables with file paths and sizes
-- Clean up intermediate files after confirmation
-
-**Must not do:**
-- Update checkpoints (verification_checklist, prd_review) before user confirmation
-- Update WORKFLOW-PROGRESS.json to completed before user confirmation
-- Change PRD document status from Draft to Confirmed before user confirmation
-- Generate completion report before user confirmation
-- Suggest next phase before user confirmation
-- Assume user silence means confirmation
-- Proceed to Phase 6.3 without explicit user confirmation
-- Delete PRD documents or clarification-related files
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: speccrew-workspace/docs/rules/agentflow-spec.md
+> Then read and execute the XML workflow in SKILL.xml block-by-block as the authoritative execution plan.

package/.speccrew/skills/speccrew-pm-requirement-analysis/SKILL.md

@@ -1,75 +1,15 @@
 ---
 name: speccrew-pm-requirement-analysis
-description: SpecCrew PM PRD Generation Skill. Generates PRD documents from clarified requirements and module design. Produces Master PRD and Sub-PRD Dispatch Plan for downstream Feature Design. Third step in PRD workflow.
+description: SpecCrew PM PRD Generation Skill. Generates PRD documents from clarified requirements and module design. Produces Master PRD and Sub-PRD Dispatch Plan for downstream Feature Design.
 tools: Read, Write, Glob, Grep
 ---
 
-# Skill Overview
-
-PRD document generation from clarification and modeling artifacts. Produces Master PRD and Sub-PRD Dispatch Plan.
-
-## Trigger Scenarios
-
-- Modeling completed (`.module-design.md` exists)
+# Trigger Scenarios
+- Modeling completed (.module-design.md exists)
 - PRD document generation needed
 - Feature breakdown and dispatch planning required
 
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `iteration_path` | string | Yes | Path to iteration directory |
-| `clarification_file` | string | No | Path to clarification summary (default: `{iteration_path}/01.product-requirement/.clarification-summary.md`) |
-| `module_design_file` | string | No | Path to module design (default: `{iteration_path}/01.product-requirement/.module-design.md`) |
-
-## Methodology Foundation
-
-Applies ISA-95 Stages 2-3 outputs for PRD generation:
-- WBS decomposition → Section 3 Functional Requirements
-- MoSCoW prioritization → Section 3.4 Feature Breakdown
-
-## PM Stage Content Boundary
-
-> **DO NOT INCLUDE:** API definitions, DB structures, code snippets, technical terminology.
-> **These belong to:** Feature Designer or System Designer.
-> **Required:** Use BUSINESS LANGUAGE only.
-
----
-
-# AgentFlow Definition
-
+## AgentFlow Definition
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
----
-
-# Output Checklist
-
-- [ ] Both prerequisite files verified
-- [ ] PRD template loaded
-- [ ] PRD structure determined (Single vs Master-Sub)
-- [ ] All sections filled with business language only
-- [ ] Feature Breakdown extracted with appropriate granularity
-- [ ] Task granularity checked (completable in one iteration)
-- [ ] File list confirmed with user
-- [ ] Master PRD written to correct path
-- [ ] [Master-Sub] Dispatch Plan generated
-- [ ] [Master-Sub] Returned control to PM Agent
-
----
-
-# Constraints
-
-**Must do:**
-- Verify both prerequisite files exist before starting
-- Use business language only in all PRD content
-- Use `search_replace` for section filling (not `create_file`)
-- Include Feature Breakdown in every PRD/Sub-PRD
-- Pass control to PM Agent after dispatch plan
-
-**Must not do:**
-- Include technical implementation details
-- Generate Sub-PRD files directly (dispatch plan only)
-- Skip user confirmation for file list
-- Proceed without prerequisite files
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: speccrew-workspace/docs/rules/agentflow-spec.md
+> Then read and execute the XML workflow in SKILL.xml block-by-block as the authoritative execution plan.

package/.speccrew/skills/speccrew-pm-requirement-assess/SKILL.md

@@ -1,107 +1,15 @@
 ---
 name: speccrew-pm-requirement-assess
-description: Quick impact assessment for new requirements. Analyzes affected modules, change type, risk level, and cross-module dependencies using system documentation and knowledge graph. Use when PM Agent needs a lightweight evaluation before deciding whether to run full requirement-analysis workflow.
+description: Quick impact assessment for new requirements. Analyzes affected modules, change type, risk level, and cross-module dependencies using system documentation and knowledge graph.
 tools: Read, Glob, Grep
 ---
 
-# Requirement Assessment for Product Manager
-
-Guide PM Agent to quickly assess new requirements using system documentation, module knowledge, and knowledge graph.
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - "Assess impact scope of new requirement"
 - "Analyze which modules are affected by requirement"
-- "Determine if requirement is new or modification"
 - "Quick requirement assessment"
-- "New requirement impact analysis"
-- "How big is this requirement?"
-
-## User
-
-PM Agent (speccrew-product-manager)
-
-## Input
-
-- New requirement description
-- system-overview.md (system documentation)
-- Related {name}-overview.md files (if specific modules identified)
-- Knowledge graph (for cross-module dependency analysis)
-
-## Output
-
-- Requirement assessment report with:
-  - Impact scope (modules affected, direct + indirect)
-  - Change type (New/Modify/Mixed/Cross-module/Config)
-  - Risk level with specific factors
-  - Recommended workflow path (simple PRD vs full ISA-95 modeling)
 
 ## AgentFlow Definition
-
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-## Assessment Report Template
-
-```markdown
-# Requirement Assessment Report
-
-## Requirement Summary
-- **Description**: [Brief description]
-- **Source**: [Who requested / Which scenario]
-
-## Assessment Result
-
-### Change Type
-[ ] New Feature  [ ] Modification  [ ] Mixed  [ ] Cross-module  [ ] Configuration
-
-### Impact Scope
-| Module | Impact Type | Evidence |
-|--------|-------------|----------|
-| {Module A} | Direct | [from system-overview / graph query] |
-| {Module B} | Indirect | [upstream dependency via graph] |
-
-### Risk Level
-[ ] Low  [ ] Medium  [ ] High
-
-**Risk Factors**:
-- [List specific risks with evidence]
-
-### Recommended Workflow
-- [ ] Simple PRD (speccrew-pm-requirement-analysis, simple path)
-- [ ] Full ISA-95 Modeling + PRD (speccrew-pm-requirement-analysis, complex path)
-- [ ] Full ISA-95 Modeling + Master-Sub PRD (complex path, multi-module)
-- [ ] Configuration change only
-
-### Key Findings from Graph Analysis
-- [Cross-module dependencies discovered]
-- [Upstream/downstream impacts identified]
-
-### Related Documentation
-- [Links to relevant module docs and graph nodes]
-```
-
-## Quick Reference
-
-### Change Type Quick Guide
-
-| Scenario | Change Type | Example |
-|----------|-------------|---------|
-| New entity + new process | New Feature | Add "Refund" entity and refund process |
-| Existing entity, new attribute | Modification | Add "priority" field to Order |
-| Existing entity, new process step | Mixed | Add "approval" step to existing order flow |
-| Changes module interface | Cross-module | Order module needs new API from Inventory |
-| Toggle/feature flag | Configuration | Enable/disable feature per tenant |
-
-## Checklist
-
-- [ ] Requirement understood and categorized
-- [ ] System overview and related module overviews loaded
-- [ ] Knowledge graph queried for cross-module dependencies
-- [ ] Change type determined
-- [ ] Impact scope mapped (direct + indirect with graph evidence)
-- [ ] Risk level assessed with specific factors
-- [ ] Workflow path recommended
-- [ ] Assessment report generated
-
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: speccrew-workspace/docs/rules/agentflow-spec.md
+> Then read and execute the XML workflow in SKILL.xml block-by-block as the authoritative execution plan.

package/.speccrew/skills/speccrew-pm-requirement-clarify/SKILL.md

@@ -4,139 +4,12 @@ description: SpecCrew PM Requirement Clarification Skill. Dual-mode support (sim
 tools: Read, Write, Glob, Grep
 ---
 
-# Skill Overview
-
-Requirement clarification with dual-mode support. Produces `.clarification-summary.md` as the interface contract.
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - User provides initial requirement description
 - PM Agent needs to clarify before PRD generation
 - Requirement complexity assessment needed
 
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `requirement_file` | string | Yes | Path to requirement document |
-| `iteration_path` | string | Yes | Path to iteration directory |
-| `complexity_hint` | enum | No | `simple` or `complex`. Auto-detect if not provided |
-| `max_clarification_rounds` | number | No | Maximum clarification rounds (default: 5) |
-
-## Methodology Foundation
-
-Applies ISA-95 Stage 1 (Domain Description) for clarification:
-- Define domain boundary, participants, glossary
-- Results flow into `.clarification-summary.md` only
-
-## PM Stage Content Boundary
-
-> **DO NOT include:** API definitions, DB structures, code snippets, technical terminology. These belong to Feature Designer or System Designer.
-
-## Templates Used
-
-| Template | Path | When to Load | Purpose |
-|----------|------|-------------|----------|
-| Clarification Questions | `templates/CLARIFICATION-QUESTIONS-TEMPLATE.md` | **Inside Loop L1** (each round) | Round-based questionnaire for user clarification |
-| Clarification Summary | `templates/CLARIFICATION-SUMMARY-TEMPLATE.md` | **After G3 guard passes** (sufficiency_checks_passed == true) | Final summary document with all Q&A and decisions |
-
-> **CRITICAL**: These two templates are used in **separate phases**. NEVER load both templates at the same time.
-> - During the clarification loop: ONLY load `CLARIFICATION-QUESTIONS-TEMPLATE.md`
-> - After ALL sufficiency checks pass: ONLY then load `CLARIFICATION-SUMMARY-TEMPLATE.md`
-
-## Execution Flow (MANDATORY sequence)
-
-```
-Step 1: Read requirement document
-Step 2: Read system knowledge (if exists)
-Step 3: Determine complexity mode (simple/complex/auto-detect)
-Step 4: Initialize loop variables
-Step 5: Clarification Loop (both simple and complex modes):
-  5a. Load CLARIFICATION-QUESTIONS-TEMPLATE.md (ONLY this template)
-  5b. Generate .clarification-questions-round-{N}.md from template
-  5c. *** HARD STOP *** Wait for user to answer questions (user-confirm event)
-  5d. Read user's answered questions file
-  5e. Perform sufficiency checks (MUST have real user answers)
-  5f. If checks fail and rounds remain → back to 5a
-  5g. If checks pass → exit loop
-Step 6: Guard Gateway G3 (sufficiency_checks_passed == true required)
-  6a. Load CLARIFICATION-SUMMARY-TEMPLATE.md (ONLY now)
-  6b. Generate .clarification-summary.md from template
-  6c. Checkpoint: clarification complete
-```
-
-> **FORBIDDEN**: Skipping Step 5c (user-confirm), loading summary template during Step 5, or generating summary without user answers.
-
----
-
-# AgentFlow Definition
-
+## AgentFlow Definition
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
----
-
-# Output Checklist
-
-- [ ] Requirement document loaded
-- [ ] System knowledge loaded (if exists)
-- [ ] Complexity mode determined
-- [ ] Loop variables initialized (`sufficiency_checks_passed`, `round_number`)
-- [ ] Clarification conducted (appropriate mode)
-- [ ] Multi-round clarification executed until ALL 4 Sufficiency Checks pass
-- [ ] Each round generates `.clarification-questions-round-{N}.md`
-- [ ] Round counter incremented after each iteration
-- [ ] Maximum 5 rounds enforced (safety valve)
-- [ ] `.clarification-summary.md` created from template
-- [ ] `.checkpoints.json` initialized via script
-
----
-
-# Constraints
-
-**Must do:**
-- Always perform at least 1 clarification round
-- **MANDATORY**: Execute multiple rounds until ALL 4 sufficiency checks pass
-- Use file-based for complex mode or 4+ questions
-- Pass all 4 Sufficiency Checks
-- Generate `.clarification-questions-round-{N}.md` for each round
-- Update `sufficiency_checks_passed` variable after each sufficiency check
-- Increment `round_number` after each round
-- Maximum 5 rounds allowed (safety valve)
-- Use `update-progress.js` for JSON files
-- After completion, return control to PM Agent for user confirmation — DO NOT auto-proceed to PRD generation
-
-**Must not do:**
-- Skip clarification based on urgency
-- Use chat for complex requirements
-- Manually write JSON files
-- Auto-proceed to Phase 4 (PRD generation) without PM Agent's user confirmation gate
-- Auto-pass sufficiency checks without actual user answers
-- **Checkpoint management is FORBIDDEN in this Skill**
-  - MUST NOT create, write, or modify `.checkpoints.json` or any checkpoint files
-  - Checkpoint write operations are the responsibility of the orchestration layer (PM Agent)
-  - After this Skill completes, PM Agent writes checkpoints ONLY after user confirmation is received
-
-### MANDATORY: User Answer Verification Rule
-
-- Worker MUST NOT auto-pass sufficiency checks without actual user answers
-- Each clarification round MUST wait for user to fill in answers before proceeding
-- Checkpoint writing is FORBIDDEN in this Skill — checkpoints are managed by the orchestration layer
-- The sufficiency check result is ONLY valid when based on real user-provided answers
-
-### CRITICAL: User Confirmation Event Enforcement
-
-1. **Event Block = Hard Stop**: Each `<block type="event" action="user-confirm">` (E-ROUND-CONFIRM) MUST trigger a pause. Worker MUST NOT proceed until user provides explicit confirmation.
-2. **No "Automated Scenario" Bypass**: Even in automated dispatch mode (speccrew-task-worker), `user-confirm` events override Worker autonomy. These are explicit user interaction gates, NOT internal automation points.
-3. **Template-Based Generation Only**: All clarification question files (`.clarification-questions-round-{N}.md`) MUST be generated using the loaded template (CLARIFICATION-QUESTIONS-TEMPLATE.md) with variable substitution. Freeform generation is FORBIDDEN.
-4. **Template-Based Summary Only**: The `.clarification-summary.md` MUST be generated using CLARIFICATION-SUMMARY-TEMPLATE.md with variable substitution. Freeform generation is FORBIDDEN.
-
-### CRITICAL: Summary Generation Guard
-
-Worker MUST strictly enforce the following before generating `.clarification-summary.md`:
-
-1. **Guard Check**: `sufficiency_checks_passed` MUST be `true` before proceeding to summary generation
-2. **Event Block Execution**: When encountering `user-confirm` event blocks (E-ROUND-CONFIRM), Worker MUST pause and wait for user confirmation. Worker MUST NOT auto-confirm or skip.
-3. **Variable State Verification**: After loop L1 exits, verify `sufficiency_checks_passed == true`. If false, do NOT generate summary — report error and exit.
-4. **FORBIDDEN**: Generating `.clarification-summary.md` when user has not answered clarification questions is a CRITICAL workflow violation.
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: speccrew-workspace/docs/rules/agentflow-spec.md
+> Then read and execute the XML workflow in SKILL.xml block-by-block as the authoritative execution plan.