speccrew 0.7.74 → 0.7.76

package/.speccrew/skills/speccrew-pm-knowledge-detector/SKILL.md

@@ -1,99 +1,15 @@
 ---
 name: speccrew-pm-knowledge-detector
-description: Detect business knowledge base availability and completeness status. Scans sync-state directory to determine if full knowledge (system-overview.md), lite knowledge (features-*.json), or no knowledge is available.
+description: Detect business knowledge base availability and completeness status. Scans sync-state directory to determine if full knowledge, lite knowledge, or no knowledge is available.
 tools: Read, Glob
 ---
 
-# Knowledge Base Detector
-
-Detect business knowledge base availability and completeness status. Scans the sync-state directory to determine the current knowledge state.
-
-## Language Adaptation
-
-**CRITICAL**: Generate all content in the language specified by the `language` parameter.
-
-- `language: "zh"` → Generate all content in 中文
-- `language: "en"` → Generate all content in English
-- Other languages → Use the specified language
-
-**All output content must be in the target language only.**
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - "Check knowledge base status"
 - "Detect if business knowledge exists"
 - "Scan knowledge availability"
-- "What knowledge is available?"
-
-## Input
-
-| Variable | Type | Description | Required |
-|----------|------|-------------|----------|
-| `workspace_path` | string | Absolute path to speccrew workspace (e.g., `speccrew-workspace`) | **Yes** |
-| `sync_state_bizs_dir` | string | Absolute path to `knowledges/base/sync-state/knowledge-bizs/` directory | **Yes** |
-| `configs_dir` | string | Absolute path to `docs/configs/` directory | **Yes** |
-
-## Output JSON
-
-```json
-{
-  "status": "full | lite | none",
-  "has_system_overview": false,
-  "has_features": false,
-  "has_entry_dirs": false,
-  "available_platforms": ["web-vue3", "backend-spring"],
-  "module_count": 0,
-  "features_files": ["path/to/features-web-vue3.json"],
-  "entry_dirs_files": ["path/to/entry-dirs-web-vue3.json"],
-  "system_overview_path": null,
-  "message": "Knowledge base status detected"
-}
-```
-
-**Status Definitions**:
-
-| Status | Condition |
-|--------|-----------|
-| `full` | has_system_overview = true AND has_features = true |
-| `lite` | has_features = true AND has_system_overview = false |
-| `none` | has_features = false AND has_system_overview = false |
 
 ## AgentFlow Definition
-
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-## Constraints
-
-1. **READ-ONLY**: This skill does not modify any files
-2. **Fast execution**: Use Glob for scanning, avoid deep file reads
-3. **Graceful handling**: Return empty arrays if directories don't exist
-4. **Path format**: All returned paths should be relative to workspace_path
-
-> **MANDATORY**: Use the provided absolute paths directly. DO NOT construct or derive paths yourself.
-
-## Task Completion Report
-
-When the task is complete, report:
-
-```json
-{
-  "status": "success | failed",
-  "skill": "speccrew-pm-knowledge-detector",
-  "detection_result": {
-    "status": "full | lite | none",
-    "available_platforms": [...],
-    "module_count": 0
-  },
-  "message": "Knowledge base status detected successfully"
-}
-```
-
-## Checklist
-
-- [ ] Step 1: Checked system-overview.md existence
-- [ ] Step 2: Scanned for features-*.json files
-- [ ] Step 3: Scanned for entry-dirs-*.json files
-- [ ] Step 4: Counted platforms and modules
-- [ ] Step 5: Determined overall status and returned JSON
+> **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-module-initializer/SKILL.md

@@ -1,189 +1,15 @@
 ---
 name: speccrew-pm-module-initializer
-description: Generate analyze task plan for a single business module. Reads features-*.json, filters matched module's pending features, and outputs a task plan JSON (list of features to analyze with analyzer parameters). Used by Worker Agent, invoked by PM Agent for on-demand module task planning.
+description: Generate analyze task plan for a single business module. Reads features-*.json, filters matched module's pending features, and outputs a task plan JSON. Used by Worker Agent, invoked by PM Agent for on-demand module task planning.
 tools: Read, Write, Skill, Bash
 ---
 
-# Module Initializer (Task Plan Generator)
-
-Generate analyze task plan for a single business module. Reads features-*.json, filters matched module's pending features, and outputs a task plan JSON (list of features to analyze with analyzer parameters). Used by Worker Agent, invoked by PM Agent for on-demand module task planning.
-
-> **Positioning**: Task plan generator for PM phase. Generates task list for PM Agent to orchestrate. Does NOT execute analyze or summarize directly — PM Agent handles Worker dispatch based on task plan output.
-
-## Language Adaptation
-
-**CRITICAL**: Generate all content in the language specified by the `language` parameter.
-
-- `language: "zh"` → Generate all content in Chinese
-- `language: "en"` → Generate all content in English
-- Other languages → Use the specified language
-
-**All output content must be in the target language only.**
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - "Generate task plan for module {name}"
 - "Plan module {name} initialization"
 - "List pending features for module {name}"
 
-## Input
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `source_path` | string | Yes | Project source code root path |
-| `module_name` | string | Yes | Module name |
-| `platform_id` | string | Yes | Platform ID (e.g., "web-vue", "admin-api") |
-| `platform_type` | string | Yes | web / mobile / backend / desktop |
-| `platform_subtype` | string | No | Platform subtype (e.g., "vue", "spring-boot") |
-| `tech_stack` | array | No | Platform tech stack (e.g., ["java", "spring-boot"]) |
-| `features_file` | string | Yes | Path to the platform's features-{platform}.json file |
-| `output_path` | string | Yes | Knowledge base output root path (e.g., speccrew-workspace/knowledges) |
-| `completed_dir` | string | Yes | Marker file output directory for api-analyze .done.json markers. Value from PM Agent: `{sync_state_bizs_dir}/completed` |
-| `sourceFile` | string | Yes | Features JSON filename (e.g., "features-backend-fastapi.json"), used for api-analyze marking |
-| `language` | string | Yes | Output language (zh / en) |
-| `workspace_path` | string | Yes | Workspace root path for constructing absolute paths |
-
-## Output JSON
-
-```json
-{
-  "module_name": "system",
-  "platform_id": "web-vue",
-  "analyzer_skill": "speccrew-knowledge-bizs-ui-analyze",
-  "tasks": [
-    {
-      "id": "ki-web-vue-system-index",
-      "fileName": "index",
-      "sourcePath": "src/views/system/user/index.vue",
-      "documentPath": "knowledges/bizs/web-vue/system/features",
-      "module": "system",
-      "platform_type": "web",
-      "platform_subtype": "vue",
-      "tech_stack": ["vue", "typescript"],
-      "language": "zh",
-      "status": "pending"
-    }
-  ],
-  "total_pending": 90,
-  "summarize_params": {
-    "skill": "speccrew-knowledge-module-summarize",
-    "module_name": "system",
-    "module_path": "knowledges/bizs/web-vue/system",
-    "language": "zh"
-  }
-}
-```
-
-**Field Definitions**:
-
-| Field | Description |
-|-------|-------------|
-| `module_name` | Module being processed |
-| `platform_id` | Platform identifier |
-| `analyzer_skill` | Selected analyzer skill name |
-| `tasks` | Array of pending features to analyze |
-| `total_pending` | Count of pending features |
-| `summarize_params` | Parameters for module-summarize skill (to be executed after all analyze tasks complete) |
-
-**Task ID Naming Convention**:
-
-The `id` field follows the format: `ki-{platform_id}-{module}-{fileName}`
-- `ki` = knowledge initialization
-- `platform_id` = Platform identifier (e.g., "web-vue", "admin-api")
-- `module` = Business module name
-- `fileName` = Feature file name
-
-Example: `"ki-web-vue-system-index"`
-
-> **Note**: This ID is consistent with task IDs in DISPATCH-PROGRESS.json and is used for progress tracking.
-
 ## AgentFlow Definition
-
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-## Constraints
-
-1. **Output path format**: Must match bizs-dispatch format: `{output_path}/bizs/{platform_id}/{module_name}/`
-
-2. **Use same analyzers as dispatch**: 
-   - Backend → `speccrew-knowledge-bizs-api-analyze`
-   - Web/Mobile/Desktop → `speccrew-knowledge-bizs-ui-analyze`
-
-3. **Worker context**: This Skill runs in Worker Agent context, invoked by PM Agent
-
-4. **NO execution**: This Skill generates task plan ONLY. It does NOT:
-   - Execute analyzer skills
-   - Execute summarize skills
-   - Update features.json analyzed field
-   - All execution is handled by PM Agent based on task plan output
-
-5. **Mutual exclusion with full dispatch**: Do not run simultaneously with full dispatch process
-
-## Error Handling
-
-| Scenario | Action |
-|----------|--------|
-| Features file not found | Return error with message |
-| No features for module | Return empty tasks array with summarize_params |
-| No pending features | Return empty tasks array with summarize_params |
-
-## Reference: Analyzer Input Parameters
-
-### API Analyzer (speccrew-knowledge-bizs-api-analyze)
-
-| Variable | Type | Description | Example |
-|----------|------|-------------|---------|
-| `fileName` | string | Controller file name | `"UserController"` |
-| `sourcePath` | string | Relative path to source file | `"src/.../UserController.java"` |
-| `documentPath` | string | Target path for generated document | `"knowledges/bizs/.../features"` |
-| `module` | string | Business module name | `"system"` |
-| `analyzed` | boolean | Analysis status flag | `false` |
-| `platform_type` | string | Platform type | `"backend"` |
-| `platform_subtype` | string | Platform subtype | `"spring-boot"` |
-| `tech_stack` | array | Platform tech stack | `["java", "spring-boot"]` |
-| `language` | string | Target language | `"zh"` |
-| `completed_dir` | string | Marker output directory (absolute path) | `".../knowledges/base/sync-state/completed"` |
-| `sourceFile` | string | Source features JSON filename | `"features-admin-api.json"` |
-
-### UI Analyzer (speccrew-knowledge-bizs-ui-analyze)
-
-| Variable | Type | Description | Example |
-|----------|------|-------------|---------|
-| `fileName` | string | Feature file name | `"index"` |
-| `sourcePath` | string | Relative path to source file | `"src/views/system/user/index.vue"` |
-| `documentPath` | string | Target path for generated document | `"knowledges/bizs/.../features"` |
-| `module` | string | Business module name | `"system"` |
-| `analyzed` | boolean | Analysis status flag | `false` |
-| `platform_type` | string | Platform type | `"web"` |
-| `platform_subtype` | string | Platform subtype | `"vue"` |
-| `tech_stack` | array | Platform tech stack | `["vue", "typescript"]` |
-| `language` | string | Target language | `"zh"` |
-
-## Task Completion Report
-
-When the task is complete, report:
-
-```json
-{
-  "status": "success",
-  "skill": "speccrew-pm-module-initializer",
-  "output": {
-    "module_name": "...",
-    "platform_id": "...",
-    "analyzer_skill": "...",
-    "tasks": [...],
-    "total_pending": <count>,
-    "summarize_params": {...}
-  }
-}
-```
-
-## Checklist
-
-- [ ] Step 1: Read features file and filtered by module
-- [ ] Step 2: Selected analyzer type based on platform_type
-- [ ] Step 3: Built task list with analyzer parameters
-- [ ] Step 4: Built summarize parameters
-- [ ] Step 5: Output task plan JSON
+> **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-module-matcher/SKILL.md

@@ -4,111 +4,12 @@ description: Match user requirement text against business knowledge base feature
 tools: Read
 ---
 
-# Module Matcher
-
-Match user requirement text against business knowledge base features to identify related modules. Performs semantic matching based on keywords, module names, and feature definitions.
-
-## Language Adaptation
-
-**CRITICAL**: Generate all content in the language specified by the `language` parameter.
-
-- `language: "zh"` → Generate all content in 中文
-- `language: "en"` → Generate all content in English
-- Other languages → Use the specified language
-
-**All output content must be in the target language only.**
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - "Find modules for requirement X"
 - "Which modules handle user management?"
 - "Match requirement to knowledge base"
-- "Identify relevant modules"
-
-## Input
-
-| Variable | Type | Description | Required |
-|----------|------|-------------|----------|
-| `requirement_text` | string | User requirement text to match | **Yes** |
-| `features_files` | string[] | Array of paths to features-*.json files | **Yes** |
-| `language` | string | Target language for output | **Yes** |
-
-## Output JSON
-
-```json
-{
-  "matched_modules": [
-    {
-      "platform_id": "web-vue3",
-      "module_name": "system",
-      "platform_type": "web",
-      "confidence": "high | medium | low",
-      "matching_features": ["user-management", "role-management"],
-      "feature_count": 15,
-      "analyzed_count": 8,
-      "source_path": "src/views/system"
-    }
-  ],
-  "unmatched_keywords": ["workflow", "approval"],
-  "recommendation": "Suggested modules based on requirement analysis",
-  "total_platforms_scanned": 3,
-  "total_modules_scanned": 12,
-  "message": "Matching completed"
-}
-```
-
-**Confidence Levels**:
-
-| Level | Condition |
-|-------|-----------|
-| `high` | Direct keyword match with module name or feature fileName |
-| `medium` | Partial match or synonym match |
-| `low` | Related concept match only |
 
 ## AgentFlow Definition
-
 <!-- @agentflow: SKILL.xml -->
-
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-## Constraints
-
-1. **READ-ONLY**: This skill does not modify any files
-2. **Handle Chinese+English**: Support bilingual text matching
-3. **Skip unreadable files**: Continue processing even if some files fail
-4. **No external API**: Matching is done locally without LLM calls
-
-## Error Handling
-
-| Scenario | Action |
-|----------|--------|
-| File not found | Skip file, log warning |
-| Invalid JSON | Skip file, log error |
-| Empty features array | Continue with empty module list |
-| No keywords extracted | Return empty matches with warning |
-
-## Task Completion Report
-
-When the task is complete, report:
-
-```json
-{
-  "status": "success | partial | failed",
-  "skill": "speccrew-pm-module-matcher",
-  "matching_result": {
-    "matched_count": 3,
-    "high_confidence_count": 1,
-    "medium_confidence_count": 1,
-    "low_confidence_count": 1
-  },
-  "message": "Module matching completed"
-}
-```
-
-## Checklist
-
-- [ ] Step 1: Parsed requirement text and extracted keywords
-- [ ] Step 2: Read all features files and built index
-- [ ] Step 3: Performed semantic matching
-- [ ] Step 4: Sorted results and generated recommendation
-- [ ] Returned complete JSON output
+> **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-phase0-init/SKILL.md

@@ -1,88 +1,15 @@
 ---
 name: speccrew-pm-phase0-init
-description: SpecCrew PM Phase 0 Initialization Skill. Handles iteration directory creation, WORKFLOW-PROGRESS.json management, checkpoint recovery, IDE detection, and path initialization. First step before any PM workflow execution.
+description: SpecCrew PM Phase 0 Initialization Skill. Handles iteration directory creation, WORKFLOW-PROGRESS.json management, checkpoint recovery, IDE detection, and path initialization.
 tools: Read, Write, Glob, Grep
 ---
 
-# Skill Overview
-
-Phase 0 initialization for PM workflow. Ensures proper workspace setup, iteration directory creation, progress tracking initialization, and resume state detection.
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - PM Agent starts a new workflow session
 - PM Agent needs to detect or create iteration directory
 - PM Agent needs to check resume state from previous session
-- PM Agent needs to initialize path context variables
-
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `user_requirement` | string | Yes | User requirement description or document path |
-| `workspace_root` | string | Yes | speccrew-workspace root directory path |
-| `source_path` | string | No | Project source root from .speccrewrc |
-| `language` | string | No | User language (auto-detected) |
-
-## Methodology Foundation
-
-Applies workspace initialization principles:
-- Iteration directory naming convention: `{number}-{type}-{name}`
-- Progress tracking via WORKFLOW-PROGRESS.json
-- Checkpoint recovery for session continuity
-
-## Templates Used
-
-| Template | Path | Purpose |
-|----------|------|---------|
-| WORKFLOW-PROGRESS | `templates/WORKFLOW-PROGRESS-TEMPLATE.json` | Standard workflow progress tracking with 7 predefined stages |
-
-## Output Deliverables
-
-| Deliverable | Path | Description |
-|-------------|------|-------------|
-| Iteration Directory | `{iterations_dir}/{iteration_name}/` | Created iteration structure |
-| WORKFLOW-PROGRESS.json | `{iteration_path}/WORKFLOW-PROGRESS.json` | Workflow progress tracker |
-| Path Variables | Context | All computed path variables |
-
----
-
-# 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
-
-- [ ] IDE environment detected
-- [ ] Path variables computed (workspace_path, iteration_path, etc.)
-- [ ] Iteration directory created or located
-- [ ] WORKFLOW-PROGRESS.json initialized or loaded
-- [ ] Checkpoint recovery logic executed
-- [ ] Resume target determined
-
----
-
-# Constraints
-
-**Must do:**
-- Follow iteration naming convention `{number}-{type}-{name}`
-- Use `update-progress.js` script for WORKFLOW-PROGRESS.json operations
-- Compute all path variables as absolute paths
-- Check for active iteration before creating new one
-- Copy requirement document to iteration's 00.docs directory
-- Create EXACTLY two subdirectories: `00.docs/` (original requirement documents) and `01.product-requirement/` (PRD and clarification outputs)
-- Initialize `WORKFLOW-PROGRESS.json` from `WORKFLOW-PROGRESS-TEMPLATE.json` template — stage names MUST match template exactly (01_prd, 02_feature_design, etc.)
-
-**Must not do:**
-- Manually create WORKFLOW-PROGRESS.json via Write/Edit tools
-- Skip checkpoint recovery check
-- Use relative paths in Worker dispatches
-- Create iteration directory without proper naming convention
-- Pre-create directories for later phases (`02.feature-design/`, `03.system-design/`, `04.development/`, `05.deployment/`, `06.system-test/`, `07.delivery/`) — each phase creates its own directory when needed
-- Invent directory names not defined in `workspace-structure.md`
-- Manually write WORKFLOW-PROGRESS.json content — MUST use template + update-progress.js script
-- Invent custom stage names (e.g., phase0_initialization, phase1_requirement_analysis) — use template-defined names only
+> **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-phase1-knowledge-check/SKILL.md

@@ -1,95 +1,15 @@
 ---
 name: speccrew-pm-phase1-knowledge-check
-description: SpecCrew PM Phase 1 Knowledge Base Availability Check Skill. Detects knowledge base status and initializes business knowledge as needed. Supports three-path routing (Full/Lite/None) with Path B deep initialization sequence.
+description: SpecCrew PM Phase 1 Knowledge Base Availability Check Skill. Detects knowledge base status and initializes business knowledge as needed. Supports three-path routing with Path B deep initialization sequence.
 tools: Read, Write, Glob, Grep, Bash
 ---
 
-# Skill Overview
-
-Knowledge base availability check for PM workflow. Detects existing knowledge status, routes to appropriate path, and initializes business knowledge when needed.
-
-## Trigger Scenarios
-
+# Trigger Scenarios
 - PM Agent starts new requirement processing
 - Knowledge base status unknown
-- Feature inventory generation needed
 - Module matching and deep initialization required
 
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `workspace_path` | string | Yes | Absolute path to speccrew-workspace root |
-| `sync_state_bizs_dir` | string | Yes | Absolute path to knowledges/base/sync-state/knowledge-bizs |
-| `configs_dir` | string | Yes | Absolute path to docs/configs directory |
-| `ide_skills_dir` | string | No | Absolute path to IDE skills directory |
-| `source_path` | string | No | Project source root from .speccrewrc |
-| `language` | string | No | User language (auto-detected if not provided) |
-| `user_requirement` | string | No | User requirement text for module matching |
-| `update_progress_script` | string | Yes | Absolute path to update-progress.js script |
-
-## Methodology Foundation
-
-Applies knowledge detection and initialization flow:
-- Stage 0: Feature inventory (lightweight metadata scan)
-- Stage 1: Deep module initialization (requirement-scoped analysis)
-
-## Output Files
-
-| File | Path | Purpose |
-|------|------|---------|
-| Features Inventory | `{sync_state_bizs_dir}/features-{platform}.json` | Platform module and feature metadata |
-| Entry Dirs | `{sync_state_bizs_dir}/entry-dirs-{platform}.json` | Source directory mapping |
-| Module Overview | `{workspace_path}/knowledges/bizs/{platform}/{module}/module-overview.md` | Module documentation |
-| System Overview | `{workspace_path}/knowledges/bizs/system-overview.md` | Aggregated system summary |
-| Graph Data | `{sync_state_bizs_dir}/completed/*.graph.json` | API/UI relationship graph data |
-
----
-
-# 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
-
-- [ ] Knowledge detector dispatched and status obtained
-- [ ] Correct path selected based on status (A/B/C)
-- [ ] **Path A**: System overview summary loaded
-- [ ] **Path B**: Module matcher executed
-- [ ] **Path B Step 1.5**: DISPATCH-PROGRESS.json initialized
-- [ ] **Path B Step 2**: Analyze task plan generated for matched modules
-- [ ] **Path B Step 3**: All feature analysis workers dispatched (parallel)
-- [ ] **Path B Step 3.5**: Graph data generated for all completed analyses
-- [ ] **Path B Step 4**: Module summaries generated
-- [ ] **Path B Step 5**: Features status updated (analyzed=true)
-- [ ] **Path B Step 6**: System overview generated
-- [ ] **Path B Step 7**: Intermediate files cleaned up
-- [ ] **Path C**: Feature inventory generated via Worker
-- [ ] **Path C Re-check**: Detector re-run and status verified
-- [ ] Knowledge context stored for downstream phases
-
----
-
-# Constraints
-
-**Must do:**
-- MUST dispatch Worker with `speccrew-pm-knowledge-detector` skill - DO NOT manually search directories
-- MUST dispatch Worker for `speccrew-knowledge-bizs-init-features` when status is "none"
-- MUST dispatch Worker for `speccrew-pm-module-matcher` in Path B
-- MUST execute all Path B Steps 1-7 in sequence - DO NOT skip to Phase 2
-- MUST use `run_in_terminal` for all script executions via Workers
-- MUST use parallel dispatch (max 5 concurrency) for feature analysis
-- MUST generate graph data (Step 3.5) before module summaries (Step 4)
-- MUST clean up intermediate files after successful completion
-
-**Must not do:**
-- Manually create features-*.json or entry-dirs-*.json files
-- Execute knowledge-base scripts directly via Bash (PM Agent must use Workers)
-- Write files to `knowledges/techs/*/sync-state/` - output goes to `knowledges/base/sync-state/knowledge-bizs/` ONLY
-- Skip Steps 2-5 when matcher returns matched modules
-- Ask user "do you want to continue?" mid-way through Path B sequence
-- Expose internal concepts (Stage 0, Stage 1) to user
+> **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.