speccrew 0.6.68 → 0.7.0

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

@@ -36,351 +36,11 @@ Applies ISA-95 Stages 2-3 outputs for PRD generation:
 
 ---
 
-# Workflow
+# AgentFlow Definition
 
-## Absolute Constraints
+<!-- @agentflow: workflow.agentflow.xml -->
 
-> **Violation = task failure**
-
-1. **FORBIDDEN:** `create_file` for documents — use `search_replace` on templates
-2. **FORBIDDEN:** Full-file rewrite — use targeted `search_replace`
-3. **MANDATORY:** Template-first workflow
-
-## Step 1: Verify Prerequisites
-
-> 🚨 **MANDATORY: Both `.clarification-summary.md` AND `.module-design.md` MUST exist.**
-> These files are generated by prior skills (clarify → model).
-
-**IF either file is missing → ABORT immediately:**
-```
-❌ ERROR: Prerequisite files not found.
-   Missing: [filename]
-   Path: {iteration_path}/01.product-requirement/
-   
-   This skill requires completed clarification and modeling.
-   Please run speccrew-pm-requirement-clarify and speccrew-pm-requirement-model first.
-```
-
-**Actions:**
-1. Read `.clarification-summary.md` — extract: key decisions, clarified scope, business rules
-2. Read `.module-design.md` — extract:
-   - **Module count**: from `## Module List (Total: N modules)` heading (parse the number N)
-   - **Module list**: rows from Module List table
-   - **Dependency matrix**: rows from Dependency Matrix table
-   - **Implementation phases**: items from Implementation Phases section
-
-   **Validation:**
-   - IF module_count >= 2 → MANDATORY: Use Master-Sub Structure
-   - IF module_count < 2 → Use Single PRD Structure
-
-## Step 2: Read PRD Template
-
-**Actions:**
-1. Search for template using glob pattern:
-   ```
-   **/speccrew-pm-requirement-analysis/templates/PRD-TEMPLATE.md
-   ```
-2. Read the template file
-
-> ⚠️ The template is ALWAYS located in the skill's own `templates/` subfolder.
-
-## Step 3: Determine PRD Structure
-
-**Structure Decision (MANDATORY IF/THEN):**
-
-```
-IF modules_count >= 2 OR cross_module_dependencies exist:
-  → MANDATORY: Use Master-Sub Structure
-  → Record: sub_prd_count = len(module_list)
-ELSE:
-  → Use Single PRD Structure
-```
-
-### Structure Types
-
-**Single PRD Structure:**
-```
-01.product-requirement/
-└── [feature-name]-prd.md
-```
-
-**Master-Sub Structure:**
-```
-01.product-requirement/
-├── [feature-name]-prd.md            # Master PRD
-├── [feature-name]-sub-[module1].md   # Sub-PRD: Module 1
-├── [feature-name]-sub-[module2].md   # Sub-PRD: Module 2
-└── ...
-```
-
-### Master PRD Requirements
-
-- Overall background, goals, and success metrics
-- System architecture overview (Mermaid graph TB)
-- Module list with scope boundaries
-- Cross-module dependency matrix
-- Implementation phases
-- Shared entities and data contracts
-- Global non-functional requirements
-
-### Sub-PRD Requirements (per module)
-
-- Module-specific user stories and functional requirements
-- Module-internal process flows and use cases
-- **Feature Breakdown** (required): Feature ID, Type, dependencies
-- Module-specific acceptance criteria
-- Interface contracts with other modules
-
-## Step 4: Write PRD Content
-
-> 🚨 **CRITICAL: PM Stage Content Boundary Enforcement**
->
-> **ABORT if ANY section includes:**
-> - API definitions, HTTP methods, JSON schemas
-> - Database table structures, ER diagrams, SQL queries
-> - Code snippets, pseudocode, implementation logic
-> - Technical terminology (UUID, JWT, REST, Microservice, etc.)
->
-> **All descriptions MUST use business language only.**
-> ✅ "Customer can upload photos" ❌ "Implement S3 upload API"
-> ✅ "Show error when name is empty" ❌ "throw ValidationException"
-
-**Actions:**
-1. Fill PRD sections using `search_replace` on template
-
-**Section Content:**
-
-| Section | Content |
-|---------|---------|
-| 1. Background & Goals | Why we're doing this, success metrics |
-| 2. User Stories | `As a [role], I want [action], so that [value]` |
-| 3. Functional Requirements | Group by priority (P0/P1/P2), include Feature Breakdown |
-| 4. Non-functional Requirements | Performance, security, compatibility |
-| 5. Acceptance Criteria | Quantifiable, verifiable |
-| 6. Boundary Description | In/out scope |
-| 7. Assumptions & Dependencies | Prerequisites, external dependencies |
-
-### 4.1 Feature Breakdown
-
-**For each module, extract business operation units:**
-
-| Feature ID | Feature Name | Type | Priority | Scope | Description |
-|------------|-------------|------|----------|-------|-------------|
-| F-{MODULE}-01 | {name} | User Interaction / Backend Process | P0/P1/P2 | {scope} | {desc} |
-
-**Feature Types:**
-- `User Interaction`: Features involving user interface and business logic
-- `Backend Process`: Background processing without direct user interaction
-
-**Granularity Guidelines:**
-| Good Size | Too Large (Split) |
-|-----------|-------------------|
-| Single CRUD operation group | Complete module with 5+ CRUDs |
-| One list page with filters | Entire reporting subsystem |
-| One form with validation | Multi-step wizard with 10+ steps |
-
-### 4.2 Section 3.5 Feature Details — Two-Phase Strategy
-
-> **⚠️ CRITICAL: Two-Phase Strategy (Skeleton-First, Content-After)**
->
-> This step MUST be executed in two phases to ensure consistent document structure.
-
-#### Phase A: Skeleton Construction (BEFORE any content filling)
-
-1. Read PRD-TEMPLATE.md to identify the complete Feature Details structure
-2. Count the number of features from `{module_features}` input
-3. For Section 3.5 Feature Details, replicate the template's Feature block structure for EACH feature:
-   - Copy the EXACT template structure (all 6 sub-sections) from PRD-TEMPLATE.md
-   - Create `#### Feature 1: {feature_name}` through `#### Feature N: {feature_name}`
-   - Each feature block MUST contain these 6 sub-section headers (copied from template):
-     ```
-     **Requirement Description:**
-     [TO BE FILLED]
-     
-     **Interaction Flow:**
-     [TO BE FILLED]
-     
-     **Boundary Conditions:**
-     [TO BE FILLED]
-     
-     **Exception Scenarios:**
-     [TO BE FILLED]
-     
-     **Operation Flow Diagram:**
-     [TO BE FILLED]
-     
-     **Operation Steps Detail:**
-     [TO BE FILLED]
-     ```
-4. Verify skeleton: confirm ALL features have ALL 6 sub-section headers before proceeding
-
-> ⚠️ DO NOT start filling content until the complete skeleton is verified.
-
-#### Phase B: Content Filling (AFTER skeleton is complete)
-
-Fill each `[TO BE FILLED]` placeholder with actual content:
-- **Requirement Description** → Business requirements in business language
-- **Interaction Flow** → User interaction steps (numbered list)
-- **Boundary Conditions** → Table with Condition Type | Scenario | Handling Rule
-- **Exception Scenarios** → Bullet list of exception handling
-- **Operation Flow Diagram** → Mermaid `graph LR` diagram showing operation flow
-- **Operation Steps Detail** → Table: Step | Action | Expected Outcome | Exception Handling
-
----
-
-### 4.3 Marking Existing vs New Features
-
-| Marker | Meaning |
-|--------|---------|
-| `[EXISTING]` | Current system capability being reused |
-| `[MODIFIED]` | Existing feature being enhanced/changed |
-| `[NEW]` | Brand new functionality |
-
-## Step 5: Task Granularity Check
-
-> **⚠️ CRITICAL: Section 3.5 Feature Details Skeleton Verification**
->
-> Before proceeding, verify the two-phase strategy was correctly applied:
-> 1. Confirm ALL features in Section 3.5 have complete 6 sub-section skeletons
-> 2. Confirm no `[TO BE FILLED]` placeholders remain unfilled
-> 3. Confirm each feature has: Requirement Description, Interaction Flow, Boundary Conditions, Exception Scenarios, Operation Flow Diagram, Operation Steps Detail
-
-**After PRD completion, verify user story granularity:**
-
-**Appropriate Granularity (completable in one iteration):**
-- Add a field and write migration
-- Add a UI component in existing page
-- Modify existing server-side logic
-- Add a filter condition to list page
-
-**Granularity Requiring Splitting:**
-- Implement a complete dashboard
-- Add a whole authentication/login system
-- Overall refactoring of existing APIs
-- Cross-domain changes involving multiple modules
-
-**IF stories need splitting → Update PRD before proceeding.**
-
-## Step 6: Determine Storage Path & Write Master PRD
-
-### 6.1 Determine Path
-
-**Single PRD Structure:**
-```
-{iteration_path}/01.product-requirement/{feature-name}-prd.md
-```
-
-**Master-Sub PRD Structure:**
-```
-{iteration_path}/01.product-requirement/
-├── {feature-name}-prd.md
-├── {feature-name}-sub-{module1-key}.md
-├── {feature-name}-sub-{module2-key}.md
-└── ...
-```
-
-### 6.2 Plan File List
-
-**For Single PRD:**
-
-| # | File | Path |
-|---|------|------|
-| 1 | PRD | {iteration}/01.product-requirement/{feature-name}-prd.md |
-
-**For Master-Sub:**
-
-| # | File Type | Module | Path |
-|---|-----------|--------|------|
-| 1 | Master PRD | (system overview) | {iteration}/01.product-requirement/{feature-name}-prd.md |
-| 2 | Sub-PRD | {module-1} | {iteration}/01.product-requirement/{feature-name}-sub-{module-1-key}.md |
-| ... | ... | ... | ... |
-
-> ⚠️ **Present file list to user for confirmation before writing.**
-
-### 6.3 Write Master PRD
-
-**For Single PRD — fill ALL sections with full detail.**
-
-**For Master-Sub — fill system-level overview ONLY:**
-
-> ⚠️ **Master PRD = System Overview ONLY. Module details go into Sub-PRDs.**
-
-- Section 1: System-wide background (2-3 paragraphs max)
-- Section 2.1: All user roles with brief descriptions
-- Section 2.2: 3-5 HIGH-LEVEL system stories (module-level capabilities)
-- Section 3.1: System-level use case diagram
-- Section 3.2: ONE cross-module end-to-end process flow
-- Section 3.3: ONE row per MODULE (not per feature)
-- Section 3.4: "See individual Sub-PRDs"
-- Section 4: System-wide NFRs (1 line each)
-- Section 5: System-wide milestones only
-- Section 8: Module list, dependency matrix, implementation phases
-- Section 9: Sub-PRD Index (all file paths)
-
-## Step 7: Generate Dispatch Plan & Return
-
-**IF Master-Sub Structure:**
-
-> ⚠️ **CRITICAL — Sub-PRD Generation Rules:**
-> 1. **DO NOT generate Sub-PRD files in this Skill** — output dispatch plan ONLY
-> 2. **DO NOT skip Sub-PRD generation** — Sub-PRDs are REQUIRED
-> 3. **RETURN control to PM Agent for Phase 4 execution**
-
-**Output Sub-PRD Dispatch Plan:**
-
-| # | Module | module_key | Sub-PRD Path | Feature Count |
-|---|--------|-----------|--------------|---------------|
-| 1 | {module-1-name} | {module-1-key} | {feature-name}-sub-{module-1-key}.md | {count} |
-| 2 | {module-2-name} | {module-2-key} | {feature-name}-sub-{module-2-key}.md | {count} |
-| ... | ... | ... | ... | ... |
-
-**For each module, provide context:**
-
-```json
-{
-  "module_name": "Module Name",
-  "module_key": "module-key",
-  "module_scope": "What this module covers",
-  "module_entities": ["Entity1", "Entity2"],
-  "module_user_stories": ["Story 1", "Story 2"],
-  "module_requirements": ["Requirement in business language"],
-  "module_features": [
-    {
-      "feature_id": "F-MOD-01",
-      "feature_name": "Feature Name",
-      "type": "User Interaction",
-      "priority": "P0",
-      "description": "Business description"
-    }
-  ],
-  "module_dependencies": ["Depends on module X for Y"]
-}
-```
-
-> ⚠️ **MANDATORY: Content Boundary Constraint for Workers**
-> "PM Stage Content Boundary applies — NO API definitions, NO database structures, NO code snippets. Use BUSINESS LANGUAGE ONLY."
-
-**Return message:**
-```
-→ MANDATORY: RETURN to PM Agent Phase 4 (Sub-PRD Worker Dispatch)
-→ PM Agent invokes speccrew-task-worker for each module
-→ Each worker uses speccrew-pm-sub-prd-generate/SKILL.md
-→ After ALL workers complete → PM Agent executes Phase 5 Verification
-```
-
-🛑 **CRITICAL HANDOFF — PM Agent MUST NOT:**
-- Generate Sub-PRD files directly
-- Invoke speccrew-pm-sub-prd-generate as its own operation
-- Create Sub-PRD content using create_file or any file write
-- Skip speccrew-task-worker and directly produce Sub-PRD documents
-
-**PM Agent MUST:**
-1. Use Dispatch Plan above to invoke speccrew-task-worker for EACH module
-2. Each worker uses skill_path: speccrew-pm-sub-prd-generate/SKILL.md
-3. Initialize DISPATCH-PROGRESS.json via update-progress.js script (NOT manually)
-4. After ALL workers complete → Execute Phase 5 Verification
-
-**IF Single PRD Structure:** Skip dispatch plan (no Sub-PRDs needed).
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
 
 ---
 

package/.speccrew/skills/speccrew-pm-requirement-analysis/workflow.agentflow.xml

@@ -0,0 +1,174 @@
+<?xml version="1.0" encoding="UTF-8"?>
+<workflow id="requirement-analysis-main" status="pending" version="1.0" desc="Generate PRD documents from clarified requirements and module design">
+
+  <!-- ============================================================
+       Input Parameters Definition
+       ============================================================ -->
+  <block type="input" id="I1" desc="Workflow input parameters">
+    <field name="iteration_path" required="true" type="string" desc="Path to iteration directory"/>
+    <field name="clarification_file" required="false" type="string" desc="Path to clarification summary (default: {iteration_path}/01.product-requirement/.clarification-summary.md)"/>
+    <field name="module_design_file" required="false" type="string" desc="Path to module design (default: {iteration_path}/01.product-requirement/.module-design.md)"/>
+    <field name="language" required="true" type="string" desc="Language code for generated content"/>
+  </block>
+
+  <!-- ============================================================
+       Global Constraints
+       ============================================================ -->
+  <block type="rule" id="R1" level="forbidden" desc="Document generation constraints">
+    <field name="text">FORBIDDEN: create_file for documents - use search_replace on templates</field>
+    <field name="text">FORBIDDEN: Full-file rewrite - use targeted search_replace</field>
+    <field name="text">MANDATORY: Template-first workflow</field>
+  </block>
+
+  <block type="rule" id="R2" level="forbidden" desc="Content boundary constraints">
+    <field name="text">DO NOT include: API definitions, DB structures, code snippets, technical terminology</field>
+    <field name="text">These belong to: Feature Designer or System Designer</field>
+    <field name="text">Required: Use BUSINESS LANGUAGE only</field>
+  </block>
+
+  <block type="rule" id="R3" level="forbidden" desc="Sub-PRD generation constraints">
+    <field name="text">DO NOT generate Sub-PRD files in this Skill - output dispatch plan ONLY</field>
+    <field name="text">DO NOT skip Sub-PRD generation - Sub-PRDs are REQUIRED</field>
+    <field name="text">RETURN control to PM Agent for Phase 4 execution</field>
+  </block>
+
+  <!-- ============================================================
+       Main Processing Sequence
+       ============================================================ -->
+  <sequence id="S1" name="PRD Generation" status="pending" desc="Generate PRD from clarification and modeling artifacts">
+
+    <!-- Step 1: Verify Prerequisites -->
+    <block type="task" id="B1" action="read-file" desc="Read clarification summary">
+      <field name="path" value="${clarification_file}"/>
+      <field name="output" var="clarification_content"/>
+    </block>
+
+    <block type="task" id="B2" action="read-file" desc="Read module design">
+      <field name="path" value="${module_design_file}"/>
+      <field name="output" var="module_design_content"/>
+    </block>
+
+    <!-- Gateway: Validate prerequisites -->
+    <block type="gateway" id="G1" mode="guard" desc="Verify both prerequisite files exist"
+           test="${clarification_content} != null AND ${module_design_content} != null"
+           fail-action="stop">
+      <field name="message">ERROR: Prerequisite files not found. Both .clarification-summary.md AND .module-design.md MUST exist. Run speccrew-pm-requirement-clarify and speccrew-pm-requirement-model first.</field>
+    </block>
+
+    <!-- Step 2: Extract module information -->
+    <block type="task" id="B3" action="analyze" desc="Extract module count and list from module design">
+      <field name="input" value="${module_design_content}"/>
+      <field name="extract">
+        - Module count from "## Module List (Total: N modules)" heading
+        - Module list from Module List table
+        - Dependency matrix from Dependency Matrix table
+        - Implementation phases from Implementation Phases section
+      </field>
+      <field name="output" var="module_info"/>
+    </block>
+
+    <!-- Step 3: Read PRD Template -->
+    <block type="task" id="B4" action="run-script" desc="Search for PRD template">
+      <field name="command">Get-ChildItem -Path "." -Recurse -Filter "PRD-TEMPLATE.md" | Where-Object { $_.FullName -like "*speccrew-pm-requirement-analysis*" } | Select-Object -First 1 -ExpandProperty FullName</field>
+      <field name="output" var="template_path"/>
+    </block>
+
+    <block type="task" id="B5" action="read-file" desc="Read PRD template">
+      <field name="path" value="${template_path}"/>
+      <field name="output" var="template_content"/>
+    </block>
+
+    <!-- Step 4: Determine PRD Structure -->
+    <block type="gateway" id="G2" mode="exclusive" desc="Determine PRD structure based on module count">
+      <branch test="${module_info.count} >= 2" name="Master-Sub Structure">
+        <block type="task" id="B6" action="analyze" desc="Set Master-Sub structure">
+          <field name="set_var" name="prd_structure" value="master-sub"/>
+          <field name="set_var" name="sub_prd_count" value="${module_info.count}"/>
+        </block>
+      </branch>
+      <branch test="${module_info.count} < 2" name="Single PRD Structure" default="true">
+        <block type="task" id="B7" action="analyze" desc="Set Single PRD structure">
+          <field name="set_var" name="prd_structure" value="single"/>
+        </block>
+      </branch>
+    </block>
+
+    <!-- Step 5: Write Master PRD -->
+    <block type="task" id="B8" action="generate" desc="Generate Master PRD content">
+      <field name="template" value="${template_content}"/>
+      <field name="fill_sections">
+        - Section 1: Background & Goals (system-wide)
+        - Section 2: User Stories (high-level)
+        - Section 3: Functional Requirements (system-level)
+        - Section 4: Non-functional Requirements (system-wide)
+        - Section 5: Acceptance Criteria
+        - Section 8: Module list, dependency matrix, implementation phases
+        - Section 9: Sub-PRD Index (if Master-Sub)
+      </field>
+      <field name="use_business_language_only">true</field>
+      <field name="output_path" value="${iteration_path}/01.product-requirement/{feature-name}-prd.md"/>
+      <field name="output" var="master_prd_path"/>
+    </block>
+
+    <!-- Checkpoint: Master PRD written -->
+    <block type="checkpoint" id="CP1" name="master-prd-written" desc="Verify Master PRD created">
+      <field name="file" value="${master_prd_path}"/>
+      <field name="verify" value="file_exists(${master_prd_path})"/>
+    </block>
+
+    <!-- Step 6: Generate Dispatch Plan (if Master-Sub) -->
+    <block type="gateway" id="G3" mode="exclusive" desc="Generate dispatch plan if Master-Sub structure">
+      <branch test="${prd_structure} == 'master-sub'" name="Generate Dispatch Plan">
+        <block type="loop" id="L1" over="${module_info.modules}" as="module" desc="Build dispatch context for each module">
+          <block type="task" id="B9" action="analyze" desc="Build module context">
+            <field name="build_context">
+              module_id: M{index}
+              module_name: ${module.name}
+              module_key: ${module.key}
+              module_scope: ${module.scope}
+              module_entities: ${module.entities}
+              module_user_stories: ${module.user_stories}
+              module_requirements: ${module.requirements}
+              module_features: ${module.features}
+              module_dependencies: ${module.dependencies}
+            </field>
+            <field name="output" var="module_context"/>
+          </block>
+        </block>
+
+        <block type="task" id="B10" action="analyze" desc="Compile dispatch plan">
+          <field name="compile">
+            dispatch_plan: Array of module contexts
+            master_prd_path: ${master_prd_path}
+            template_path: ${template_path}
+            output_base_path: ${iteration_path}/01.product-requirement
+            feature_name: {extracted from clarification}
+          </field>
+          <field name="output" var="dispatch_plan"/>
+        </block>
+
+        <block type="event" id="E1" action="log" level="info" desc="Log handoff to PM Agent">
+          <field name="message">MANDATORY: RETURN to PM Agent Phase 4 (Sub-PRD Worker Dispatch). PM Agent invokes speccrew-task-worker for each module. Each worker uses speccrew-pm-sub-prd-generate/SKILL.md. After ALL workers complete -> PM Agent executes Phase 5 Verification.</field>
+        </block>
+      </branch>
+      <branch test="${prd_structure} == 'single'" name="Single PRD - Skip Dispatch">
+        <block type="event" id="E2" action="log" level="info" desc="Log single PRD completion">
+          <field name="message">Single PRD structure - no Sub-PRDs needed. PRD generation complete.</field>
+        </block>
+      </branch>
+    </block>
+
+  </sequence>
+
+  <!-- ============================================================
+       Output Results
+       ============================================================ -->
+  <block type="output" id="O1" desc="Workflow output results">
+    <field name="master_prd_path" from="${master_prd_path}" type="string" desc="Path to generated Master PRD"/>
+    <field name="prd_structure" from="${prd_structure}" type="string" desc="PRD structure type: single or master-sub"/>
+    <field name="dispatch_plan" from="${dispatch_plan}" type="object" desc="Sub-PRD dispatch plan (if Master-Sub)"/>
+    <field name="module_count" from="${module_info.count}" type="number" desc="Number of modules"/>
+    <field name="message" value="PRD generation complete" type="string" desc="Status message"/>
+  </block>
+
+</workflow>

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

@@ -36,99 +36,11 @@ PM Agent (speccrew-product-manager)
   - Risk level with specific factors
   - Recommended workflow path (simple PRD vs full ISA-95 modeling)
 
-## Assessment Workflow
+## AgentFlow Definition
 
-### Step 1: Understand the Requirement
+<!-- @agentflow: workflow.agentflow.xml -->
 
-Analyze the new requirement:
-- What business problem does it solve?
-- What entities are involved?
-- What processes are affected?
-
-### Step 2: Load System Context
-
-#### 2.1 Read System Overview
-
-```
-speccrew-workspace/knowledges/bizs/system-overview.md
-```
-
-Identify:
-- Which business domain the requirement belongs to
-- Which modules are potentially related
-- Which business processes are involved
-
-#### 2.2 Load Related Module Overviews
-
-For each potentially affected module:
-```
-speccrew-workspace/knowledges/bizs/{module-name}/{module-name}-overview.md
-```
-
-#### 2.3 Query Knowledge Graph (for cross-module analysis)
-
-Use `speccrew-knowledge-graph-query` skill to trace dependencies:
-
-| Action | Use Case |
-|--------|----------|
-| `search` | Find entities related to requirement keywords |
-| `trace-upstream` | What depends on the affected entities? |
-| `trace-downstream` | What do the affected entities depend on? |
-| `query-nodes` | List all entities in a suspected module |
-
-### Step 3: Determine Change Type
-
-Use decision tree:
-
-```
-New Requirement
-    |
-    +-- Involves new business entity?
-    |      +-- NEW FEATURE
-    |
-    +-- Modifies core attributes of existing entity?
-    |      +-- MODIFICATION
-    |
-    +-- Adds new process steps but reuses existing entities?
-    |      +-- MIXED (New + Modify)
-    |
-    +-- Affects cross-module data flow?
-    |      +-- CROSS-MODULE CHANGE
-    |
-    +-- Can be handled by configuration?
-           +-- CONFIGURATION
-```
-
-### Step 4: Identify Impact Scope
-
-Check for impact on:
-- **Direct modules**: Modules explicitly mentioned in requirement
-- **Dependency modules**: Upstream/downstream from graph query (Step 2.3)
-- **External integrations**: Third-party systems
-- **Data entities**: New or modified entities
-
-### Step 5: Assess Risk Level
-
-| Risk Factor | High Risk Indicators |
-|-------------|---------------------|
-| Scope | Affects 3+ modules |
-| Data | Modifies core entity structure |
-| Integration | Changes external API contracts |
-| Process | Alters critical business flow |
-| State | Modifies entity state machine |
-
-### Step 6: Recommend Workflow Path
-
-Based on assessment results, recommend the appropriate next step:
-
-| Assessment Result | Recommended Path |
-|-------------------|-----------------|
-| 1 module, low risk, clear domain | Simple PRD (skip ISA-95 modeling) |
-| 2+ modules, medium risk | Full requirement-analysis with ISA-95 modeling |
-| Cross-module, high risk | Full requirement-analysis with ISA-95 modeling + Master-Sub PRD |
-| Configuration only | Direct implementation, PRD optional |
-
-### Step 7: Generate Assessment Report
+> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
 
 ## Assessment Report Template