speccrew 0.7.73 → 0.7.75

package/.speccrew/skills/speccrew-system-deployer-orchestration/SKILL.md

@@ -1,91 +1,18 @@
 ---
 name: speccrew-system-deployer-orchestration
-version: 1.0.0
 description: System Deployer 的编排调度技能，负责执行部署工作流：阶段门禁验证、IDE目录检测、准备阶段、技能调度（构建→迁移→启动→冒烟测试）、部署总结确认。
 tools: Read, Write, Glob, Grep, Bash
 ---
 
-> **⚠️ MANDATORY EXECUTION PROTOCOL — READ BEFORE EXECUTING ANY BLOCK**
->
-> **Step 1**: Load XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md` — this defines all block types and action-to-tool mappings
->
-> **Step 2**: Execute this SKILL.md's XML workflow **block by block in document order**. For EVERY block, you MUST follow this 3-step cycle:
->
-> ```
-> 🏷️ Block [ID] (action=[action]) — [desc]
-> 🔧 Tool: [which IDE tool to call]
-> ✅ Result: [output or status]
-> ```
->
-> Action-to-tool mapping:
-> - `action="run-skill"` → Invoke via **Skill tool** (pass the `<field name="skill">` value EXACTLY)
-> - `action="run-script"` → Execute via **Terminal tool** (pass the `<field name="command">` value EXACTLY)
-> - `action="dispatch-to-worker"` → Create **Task** via **Task tool** for `speccrew-task-worker`
-> - `action="read-file"` → Read via **Read tool**
-> - `action="log"` → Output message directly
-> - `action="confirm"` → Present to user and wait for response
->
-> **Step 3**: Execute ALL blocks sequentially without pausing (only stop at explicit `<event action="confirm">` blocks)
+# Trigger Scenarios
 
-# System Deployer Orchestration
-
-System Deployer 的核心编排技能，负责：
-
-1. **Stage Gate Verification** - 验证开发阶段已完成确认
-2. **IDE Directory Detection** - 检测 IDE 目录并验证部署技能存在
-3. **Preparation** - 读取开发任务记录、加载技术知识、确定项目根目录
-4. **Skill Dispatch** - 按顺序调度部署技能：build → migrate → startup → smoke-test
-5. **Deployment Summary** - 呈现部署总结并等待用户确认
-
-## Invocation Method
-
-**CRITICAL**: This skill is loaded directly by System Deployer Agent — do NOT invoke via Worker Agent.
-
-```xml
-<block type="task" action="run-skill" desc="System Deployer orchestration workflow">
-  <field name="skill">speccrew-system-deployer-orchestration</field>
-</block>
-```
-
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `workspace_path` | string | Yes | speccrew-workspace 根目录绝对路径 |
-| `iteration_dir` | string | Yes | 当前迭代目录路径 |
-| `update_progress_script` | string | Yes | update-progress.js 脚本路径 |
-
-## Output
-
-- `status` - Execution status (success / partial / failed)
-- `deployed` - Boolean indicating deployment success
-- `checkpoints` - List of passed checkpoints
-- `output_files` - List of generated/modified files
-- `summary` - Execution summary
-- `next_steps` - Suggested next actions
-
----
+- System Deployer Agent starts deployment workflow after development confirmation
+- Development stage confirmed and build-migrate-startup-smoke pipeline needs execution
+- User requests application deployment or build verification
 
 ## AgentFlow Definition
 
 <!-- @agentflow: SKILL.xml -->
 
----
-
-## CONTINUOUS EXECUTION RULES
-
-This skill MUST execute tasks continuously without unnecessary interruptions.
-
-### FORBIDDEN Interruptions
-
-1. DO NOT ask user "Should I continue?" after completing a subtask
-2. DO NOT suggest "Let me split this into batches" or "Let's do this in parts"
-3. DO NOT pause to list what you plan to do next — just do it
-4. DO NOT ask for confirmation before invoking skills (Phase 3 HARD STOP is the only confirmation point)
-
-### When to Pause (ONLY these cases)
-
-1. Phase 3 HARD STOP — user confirmation required by design
-2. Ambiguous requirements that genuinely need clarification
-3. Unrecoverable errors that prevent further progress
-4. Skill invocation failure — report and wait for user decision
+> **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-system-designer-orchestration/SKILL.md

@@ -10,42 +10,9 @@ tools: Read, Write, Glob, Grep, Bash, Agent
 - Feature Spec and API Contract have been confirmed (02_feature_design stage = confirmed)
 - User requests system design execution for confirmed features
 
-## EXECUTION PROTOCOL
-
-### Action-to-Tool Mapping
-
-| XML Block Action | IDE Tool | How to Execute |
-|---|---|---|
-| `action="run-script"` | **Terminal tool** | Execute command in terminal |
-| `action="dispatch-to-worker"` | **Agent tool** | Create new `speccrew-task-worker` agent session (NOT Skill tool) |
-| `action="read-file"` | **Read tool** | Read file at specified path |
-| `action="log"` | **Output** | Log message directly to conversation |
-| `action="confirm"` | **Output + Wait** | Present message and wait for user response |
-
-### Worker Dispatch Protocol
-
-When executing any block with `action="dispatch-to-worker"`:
-
-1. **Use Agent tool** to create a new `speccrew-task-worker` agent
-2. Pass `skill_path` from the XML `<field name="skill_path">` to worker
-3. Pass all context parameters from XML `<field name="context">` block
-4. **Wait** for worker agent to complete before proceeding to next block
-5. Verify worker's output file exists at expected path
-
-**CRITICAL DISTINCTION**:
-- **Agent tool** = creates a NEW autonomous worker session (for dispatch-to-worker)
-- **Skill tool** = executes skill inline in current session (for run-skill)
-- Using the WRONG tool will cause dispatch failure
-
-### Phase 5 Batch Dispatch
-
-- Dispatch ONE worker per Feature×Platform combination via Agent tool
-- Max concurrent workers per batch: defined in XML
-- Pass `skip_confirmation: true` and `skip_index_generation: true` in batch context
-- Wait for all workers in current batch to complete before dispatching next batch
-
 ## AgentFlow Definition
 
 <!-- @agentflow: SKILL.xml -->
 
-> **REQUIRED**: Read and execute the XML workflow above. The XML contains the complete workflow definition including all phases (Stage Gate → Preparation → Resource Verification → Framework Evaluation → Design Overview → Platform Dispatch → Joint Confirmation), rules, conditions, and checklist.
+> **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-system-developer-orchestration/SKILL.md

@@ -1,110 +1,18 @@
 ---
 name: speccrew-system-developer-orchestration
-version: 1.0.0
 description: System Developer 的核心编排技能，负责读取系统设计蓝图、协调跨平台开发任务分发、加载技术知识、验证环境就绪、分发各平台开发技能、执行集成检查、交付开发完成报告。支持 web、mobile、desktop 和 backend 平台。
 tools: Read, Write, Glob, Grep, Bash
 ---
 
-> **⚠️ MANDATORY EXECUTION PROTOCOL — READ BEFORE EXECUTING ANY BLOCK**
->
-> **Step 1**: Load XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md` — this defines all block types and action-to-tool mappings
->
-> **Step 2**: Execute this SKILL.md's XML workflow **block by block in document order**. For EVERY block, you MUST follow this 3-step cycle:
->
-> ```
-> 🏷️ Block [ID] (action=[action]) — [desc]
-> 🔧 Tool: [which IDE tool to call]
-> ✅ Result: [output or status]
-> ```
->
-> Action-to-tool mapping:
-> - `action="run-skill"` → Invoke via **Skill tool** (pass the `<field name="skill">` value EXACTLY)
-> - `action="run-script"` → Execute via **Terminal tool** (pass the `<field name="command">` value EXACTLY)
-> - `action="dispatch-to-worker"` → Create **Task** via **Task tool** for `speccrew-task-worker`
-> - `action="read-file"` → Read via **Read tool**
-> - `action="log"` → Output message directly
-> - `action="confirm"` → Present to user and wait for response
->
-> **Step 3**: Execute ALL blocks sequentially without pausing (only stop at explicit `<event action="confirm">` blocks)
+# Trigger Scenarios
 
-# System Developer Orchestration
-
-System Developer 的核心编排技能，负责：
-
-1. **Stage Gate** - 验证 System Design 阶段已确认
-2. **Read System Design** - 读取设计文档，识别平台模块
-3. **Load Techs Knowledge** - 加载平台技术栈和 API Contract
-4. **Environment Pre-check** - 验证运行时、依赖、服务
-5. **Dispatch Dev Workers** - 分发开发任务给 Worker
-6. **Integration Check** - 验证跨平台 API 和数据一致性
-7. **Delivery Report** - 交付报告和确认
-
-## Invocation Method
-
-**CRITICAL**: This skill is loaded directly by System Developer Agent — do NOT invoke via Worker Agent.
-
-```xml
-<block type="task" action="run-skill" desc="System Developer orchestration workflow">
-  <field name="skill">speccrew-system-developer-orchestration</field>
-</block>
-```
-
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `workspace_path` | string | Yes | speccrew-workspace root directory path |
-| `iterations_dir` | string | Yes | iterations directory path |
-| `update_progress_script` | string | Yes | Path to update-progress.js script |
-| `current_iteration` | string | No | Current active iteration identifier |
-
-## Output
-
-- `status` - Execution status (success / partial / failed)
-- `output_files` - List of generated source files and task records
-- `summary` - Execution summary
-- `next_steps` - Suggested next actions
-
----
+- System Developer Agent starts development workflow after system design confirmation
+- System Design stage confirmed and cross-platform development tasks need dispatching
+- User requests code implementation for confirmed system designs
 
 ## AgentFlow Definition
 
 <!-- @agentflow: SKILL.xml -->
 
----
-
-## CONTINUOUS EXECUTION RULES
-
-This skill MUST execute tasks continuously without unnecessary interruptions.
-
-### FORBIDDEN Interruptions
-
-1. DO NOT ask user "Should I continue?" after completing a subtask
-2. DO NOT suggest "Let me split this into batches" or "Let's do this in parts"
-3. DO NOT pause to list what you plan to do next — just do it
-4. DO NOT ask for confirmation before generating output files
-5. DO NOT warn about "large number of files" — proceed with generation
-6. DO NOT offer "Should I proceed with the remaining items?"
-
-### When to Pause (ONLY these cases)
-
-1. CHECKPOINT gates defined in workflow (user confirmation required by design)
-2. Ambiguous requirements that genuinely need clarification
-3. Unrecoverable errors that prevent further progress
-4. Security-sensitive operations (e.g., deleting existing files)
-
-### CRITICAL CONSTRAINT
-
-**This skill is a pure orchestrator/dispatcher. It MUST NOT:**
-- Create source code files (*.java, *.vue, *.ts, *.py, *.dart, etc.)
-- Invoke dev skills directly (only via speccrew-task-worker)
-- Write implementation code in any language
-- Create code as fallback if worker fails
-
-### HARD STOP Checkpoints
-
-This workflow has **mandatory HARD STOP** checkpoints at:
-- **Phase 4.2a**: Task list review confirmation
-- **Phase 6.6.5**: Delivery report confirmation
-
-DO NOT proceed past these checkpoints without explicit user confirmation.
+> **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-task-worker-execution/SKILL.md

@@ -1,106 +1,18 @@
 ---
 name: speccrew-task-worker-execution
-version: 1.0.0
 description: Task Worker 的通用任务执行技能，负责接收任务参数、加载 Skill 定义、读取 AgentFlow 规范、逐 Block 执行工作流并输出结构化完成报告。
 tools: Read, Grep, Glob, Write, Bash, Edit, WebFetch, WebSearch
 ---
 
-> **⚠️ MANDATORY EXECUTION PROTOCOL — READ BEFORE EXECUTING ANY BLOCK**
->
-> **Step 1**: Load XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md` — this defines all block types and action-to-tool mappings
->
-> **Step 2**: Execute this SKILL.md's XML workflow **block by block in document order**. For EVERY block, you MUST follow this 3-step cycle:
->
-> ```
-> 🏷️ Block [ID] (type=[type], action=[action]) — [desc]
-> 🔧 Tool: [which IDE tool to call]
-> ✅ Result: [output or status]
-> ```
->
-> Action-to-tool mapping:
-> - `action="run-skill"` → Invoke via **Skill tool** (pass the `<field name="skill">` value EXACTLY)
-> - `action="run-script"` → Execute via **Terminal tool** (pass the `<field name="command">` value EXACTLY)
-> - `action="dispatch-to-worker"` → Create **Task** via **Task tool** for `speccrew-task-worker`
-> - `action="read-file"` → Read via **Read tool**
-> - `action="log"` → Output message directly
-> - `action="confirm"` → Present to user and wait for response
->
-> **Step 3**: Execute ALL blocks sequentially without pausing (only stop at explicit `<event action="confirm">` blocks)
+# Trigger Scenarios
 
-# Task Worker Execution
-
-通用任务执行 Worker 的核心执行技能，负责：
-
-1. **Task Reception** - 接收来自调用 Agent 的任务参数
-2. **Skill Discovery** - 解析技能路径，定位 SKILL.md 文件
-3. **Workflow Execution** - 加载 AgentFlow 规范，逐 Block 执行工作流
-4. **Result Reporting** - 输出结构化完成报告
-
-## Invocation Method
-
-**CRITICAL**: This skill is loaded directly by Task Worker Agent — invoked when `skill_path` or `skill_name` is provided.
-
-```xml
-<block type="task" action="run-skill" desc="Execute task worker workflow">
-  <field name="skill">speccrew-task-worker-execution</field>
-</block>
-```
-
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `skill_path` | string | No | Full relative path to SKILL.md file (priority over skill_name) |
-| `skill_name` | string | No | Skill name identifier for discovery (backward compatibility) |
-| `context` | object | Yes | Task context parameters (task_id, module, input_path, output_path, etc.) |
-| `workspace_root` | string | Yes | Workspace root directory path |
-
-## Output
-
-- `status` - Execution status (success / failed)
-- `task_id` - Task identifier from dispatch context
-- `output_files` - List of generated/modified files
-- `summary` - Execution summary
-- `error` - Error description (if failed)
-- `error_category` - Error category (DEPENDENCY_MISSING | BUILD_FAILURE | VALIDATION_ERROR | RUNTIME_ERROR | BLOCKED)
-
----
+- Orchestration Agent dispatches a task via dispatch-to-worker action
+- Task Worker Agent receives skill_path or skill_name to execute
+- User explicitly requests execution of a specific skill workflow
 
 ## AgentFlow Definition
 
 <!-- @agentflow: SKILL.xml -->
 
----
-
-## CONTINUOUS EXECUTION RULES
-
-This skill MUST execute tasks continuously without unnecessary interruptions.
-
-### FORBIDDEN Interruptions
-
-1. DO NOT ask user "Should I continue?" after completing a subtask
-2. DO NOT suggest "Let me split this into batches" or "Let's do this in parts"
-3. DO NOT pause to list what you plan to do next — just do it
-4. DO NOT ask for confirmation before generating output files
-5. DO NOT warn about "large number of files" — proceed with generation
-6. DO NOT offer "Should I proceed with the remaining items?"
-
-### When to Pause (ONLY these cases)
-
-1. CHECKPOINT gates defined in workflow (user confirmation required by design)
-2. Ambiguous requirements that genuinely need clarification
-3. Unrecoverable errors that prevent further progress
-4. Security-sensitive operations (e.g., deleting existing files)
-
-### Batch Execution Behavior
-
-- When multiple items need processing, process ALL of them sequentially without asking
-- Use DISPATCH-PROGRESS.json to track progress, enabling resumption if interrupted
-- NEVER voluntarily stop mid-batch to ask if user wants to continue
-
-### Worker Completion Protocol
-
-- After completing assigned skill execution, report results immediately
-- DO NOT ask the dispatching agent for further instructions
-- DO NOT wait for confirmation before writing output files
-- If skill execution fails, report failure with details — do not ask user what to do
+> **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-team-leader-routing/SKILL.md

@@ -1,91 +1,18 @@
 ---
 name: speccrew-team-leader-routing
-version: 1.0.0
 description: Team Leader 的意图识别与路由调度技能，负责识别用户意图并路由到对应 Skill/Agent。处理项目初始化、知识库同步、工作流诊断等基础设施请求。
 tools: Read, Write, Bash, Skill, Task
 ---
 
-> **⚠️ MANDATORY EXECUTION PROTOCOL — READ BEFORE EXECUTING ANY BLOCK**
->
-> **Step 1**: Load XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md` — this defines all block types and action-to-tool mappings
->
-> **Step 2**: Execute this SKILL.md's XML workflow **block by block in document order**. For EVERY block, you MUST follow this 3-step cycle:
->
-> ```
-> 🏷️ Block [ID] (action=[action]) — [desc]
-> 🔧 Tool: [which IDE tool to call]
-> ✅ Result: [output or status]
-> ```
->
-> Action-to-tool mapping:
-> - `action="run-skill"` → Invoke via **Skill tool** (pass the `<field name="skill">` value EXACTLY)
-> - `action="run-script"` → Execute via **Terminal tool** (pass the `<field name="command">` value EXACTLY)
-> - `action="dispatch-to-worker"` → Create **Task** via **Task tool** for `speccrew-task-worker`
-> - `action="read-file"` → Read via **Read tool**
-> - `action="log"` → Output message directly
-> - `action="confirm"` → Present to user and wait for response
->
-> **Step 3**: Execute ALL blocks sequentially without pausing (only stop at explicit `<event action="confirm">` blocks)
+# Trigger Scenarios
 
-# Team Leader Routing
-
-Team Leader 的核心路由调度技能，负责：
-
-1. **Pipeline Progress Awareness** - 读取 WORKFLOW-PROGRESS.json，感知当前流水线状态
-2. **Auto-Orchestration** - 处理自动推进与新用户引导
-3. **Intent Recognition** - 识别用户意图并路由到正确的 Skill/Agent
-4. **Skill Invocation** - 加载并执行对应的 Skill
-
-## Invocation Method
-
-**CRITICAL**: This skill is loaded directly by Team Leader Agent — do NOT invoke via Worker Agent.
-
-```xml
-<block type="task" action="run-skill" desc="Team Leader routing workflow">
-  <field name="skill">speccrew-team-leader-routing</field>
-</block>
-```
-
-## Input Parameters
-
-| Parameter | Type | Required | Description |
-|-----------|------|----------|-------------|
-| `user_message` | string | Yes | Original user input message |
-| `workspace_root` | string | Yes | speccrew-workspace root directory path |
-| `iteration_dir` | string | No | Current active iteration directory |
-
-## Output
-
-- `status` - Execution status (success / partial / failed)
-- `skill_invoked` - Name of invoked Skill (if any)
-- `output_files` - List of generated/modified files
-- `summary` - Execution summary
-- `next_steps` - Suggested next actions
-
----
+- Team Leader Agent needs to route user intent to the correct Skill or Agent
+- User submits a message requiring intent recognition and dispatch
+- Auto-orchestration or new user onboarding request
 
 ## AgentFlow Definition
 
 <!-- @agentflow: SKILL.xml -->
 
----
-
-## CONTINUOUS EXECUTION RULES
-
-This skill MUST execute tasks continuously without unnecessary interruptions.
-
-### FORBIDDEN Interruptions
-
-1. DO NOT ask user "Should I continue?" after completing a subtask
-2. DO NOT suggest "Let me split this into batches" or "Let's do this in parts"
-3. DO NOT pause to list what you plan to do next — just do it
-4. DO NOT ask for confirmation before generating output files
-5. DO NOT warn about "large number of files" — proceed with generation
-6. DO NOT offer "Should I proceed with the remaining items?"
-
-### When to Pause (ONLY these cases)
-
-1. CHECKPOINT gates defined in workflow (user confirmation required by design)
-2. Ambiguous requirements that genuinely need clarification
-3. Unrecoverable errors that prevent further progress
-4. Security-sensitive operations (e.g., deleting existing 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-test-case-design/SKILL.md

@@ -14,61 +14,5 @@ tools: Read, Write, Glob, Grep
 
 <!-- @agentflow: SKILL.xml -->
 
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-# Key Rules
-
-| Rule | Description |
-|------|-------------|
-| **No Code Implementation** | Do NOT write actual test code, only test case specifications |
-| **Coverage First** | Prioritize acceptance criteria coverage over exhaustive testing |
-| **Clear Expected Results** | Every test case must have unambiguous expected result |
-| **Traceability Required** | All test cases must trace back to requirements |
-| **Naming Convention** | Strictly follow TC-{MODULE}-{SEQ} format |
-| **Priority Alignment** | Test case priority should align with requirement priority |
-
-# Checklist
-
-- [ ] All acceptance criteria from Feature Spec have corresponding test cases
-- [ ] Each test dimension has at least one test case (if applicable)
-- [ ] Test case IDs follow naming convention (TC-{MODULE}-{SEQ})
-- [ ] All test cases have clear expected results
-- [ ] Coverage traceability matrix is complete
-- [ ] Test data sets cover normal, boundary, and exception scenarios
-- [ ] Priority assignments are consistent with requirement priorities
-- [ ] Preconditions are clearly stated for each test case
-- [ ] Steps are detailed enough for execution without ambiguity
-- [ ] Document written to correct output path
-
----
-
-# Task Completion Report
-
-Upon completion (success or failure), output the following report format:
-
-## Success Report
-```
-## Task Completion Report
-- **Status**: SUCCESS
-- **Task ID**: <from dispatch context, e.g., "test-case-web-vue">
-- **Platform**: <platform_id, e.g., "web-vue">
-- **Phase**: test_case_design
-- **Output Files**:
-  - `speccrew-workspace/iterations/{iteration}/06.system-test/cases/{platform_id}/[feature]-test-cases.md`
-- **Summary**: Test case design completed with {count} test cases covering {dimensions} dimensions
-```
-
-## Failure Report
-```
-## Task Completion Report
-- **Status**: FAILED
-- **Task ID**: <from dispatch context>
-- **Platform**: <platform_id>
-- **Phase**: test_case_design
-- **Output Files**: None
-- **Summary**: Test case design failed during {step}
-- **Error**: <detailed error description>
-- **Error Category**: DEPENDENCY_MISSING | BUILD_FAILURE | VALIDATION_ERROR | RUNTIME_ERROR | BLOCKED
-- **Partial Outputs**: <list of partially generated files or "None">
-- **Recovery Hint**: <suggestion for recovery, e.g., "Check feature spec document exists at specified path">
-```
+> **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-test-code-gen/SKILL.md

@@ -14,64 +14,5 @@ tools: Read, Write, Glob, Grep
 
 <!-- @agentflow: SKILL.xml -->
 
-> **REQUIRED**: Before executing this workflow, read the XML workflow specification: `speccrew-workspace/docs/rules/agentflow-spec.md`
-
-# Key Rules
-
-| Rule | Description |
-|------|-------------|
-| **TC ID Traceability** | Every test function MUST have a TC ID comment |
-| **Checkpoint Required** | Must confirm code plan with user before generating |
-| Convention Compliance | Follow platform-specific conventions-unit-test.md |
-| **Arrange-Act-Assert** | Maintain clear test structure |
-| **Mock Strategy Documented** | All mocks documented in code plan |
-| **No Test Execution** | This skill only generates code, does not run tests |
-
-# Checklist
-
-- [ ] Test case document read, all cases parsed
-- [ ] Technical conventions loaded (conventions-unit-test.md or inferred)
-- [ ] System design read, dependencies identified
-- [ ] Code plan generated with file-to-case mapping
-- [ ] Checkpoint passed: code plan confirmed with user
-- [ ] Every test case ID has a corresponding test function
-- [ ] Each test function has a TC ID annotation comment
-- [ ] Mock/stub strategy covers all external dependencies
-- [ ] Test code follows platform conventions-unit-test.md style
-- [ ] Shared fixtures and helpers are extracted properly
-- [ ] Arrange-Act-Assert structure maintained in tests
-- [ ] Code plan document written to correct path
-
----
-
-# Task Completion Report
-
-Upon completion (success or failure), output the following report format:
-
-## Success Report
-```
-## Task Completion Report
-- **Status**: SUCCESS
-- **Task ID**: <from dispatch context, e.g., "test-code-web-vue">
-- **Platform**: <platform_id, e.g., "web-vue">
-- **Phase**: test_code_gen
-- **Output Files**:
-  - `speccrew-workspace/iterations/{iteration}/06.system-test/code/{platform_id}/[feature]-test-code-plan.md`
-  - <list of generated test source files>
-- **Summary**: Test code generation completed with {file_count} files covering {case_count} test cases
-```
-
-## Failure Report
-```
-## Task Completion Report
-- **Status**: FAILED
-- **Task ID**: <from dispatch context>
-- **Platform**: <platform_id>
-- **Phase**: test_code_gen
-- **Output Files**: <list of partial outputs or "None">
-- **Summary**: Test code generation failed during {step}
-- **Error**: <detailed error description>
-- **Error Category**: DEPENDENCY_MISSING | BUILD_FAILURE | VALIDATION_ERROR | RUNTIME_ERROR | BLOCKED
-- **Partial Outputs**: <list of partially generated files or "None">
-- **Recovery Hint**: <suggestion for recovery, e.g., "Verify test case document format is valid">
-```
+> **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.