workflow-tdd-plan

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Workflow TDD Plan

TDD工作流规划

6-phase TDD planning workflow that orchestrates session discovery, context gathering, test coverage analysis, conflict resolution, and TDD task generation to produce implementation plans with Red-Green-Refactor cycles. Includes post-execution TDD compliance verification.
包含会话发现、上下文收集、测试覆盖率分析、冲突解决及TDD任务生成的6阶段TDD规划工作流,可生成带有红-绿-重构周期的实施计划,同时支持执行后的TDD合规验证。

Architecture Overview

架构概述

┌──────────────────────────────────────────────────────────────────┐
│  Workflow TDD Plan Orchestrator (SKILL.md)                       │
│  → Pure coordinator: Execute phases, parse outputs, pass context │
└───────────────┬──────────────────────────────────────────────────┘
   ┌────────────┼────────────┬────────────┬────────────┐
   ↓            ↓            ↓            ↓            ↓
┌────────┐ ┌────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐
│ Phase 1│ │ Phase 2│ │ Phase 3  │ │ Phase 4  │ │ Phase 5  │
│Session │ │Context │ │Test Covg │ │Conflict  │ │TDD Task  │
│Discover│ │Gather  │ │Analysis  │ │Resolve   │ │Generate  │
│ (ext)  │ │ (ext)  │ │ (local)  │ │(ext,cond)│ │ (local)  │
└────────┘ └────────┘ └──────────┘ └──────────┘ └──────────┘
    ↓           ↓           ↓            ↓            ↓
 sessionId  contextPath  testContext  resolved    IMPL_PLAN.md
            conflict_risk             artifacts   task JSONs

Phase 6: TDD Structure Validation (inline in SKILL.md)

Post-execution verification:
┌──────────────┐     ┌───────────────────┐
│  TDD Verify  │────→│ Coverage Analysis │
│  (local)     │     │ (local)           │
└──────────────┘     └───────────────────┘
phases/03-tdd-        phases/04-tdd-
verify.md             coverage-analysis.md
┌──────────────────────────────────────────────────────────────────┐
│  Workflow TDD Plan Orchestrator (SKILL.md)                       │
│  → Pure coordinator: Execute phases, parse outputs, pass context │
└───────────────┬──────────────────────────────────────────────────┘
   ┌────────────┼────────────┬────────────┬────────────┐
   ↓            ↓            ↓            ↓            ↓
┌────────┐ ┌────────┐ ┌──────────┐ ┌──────────┐ ┌──────────┐
│ Phase 1│ │ Phase 2│ │ Phase 3  │ │ Phase 4  │ │ Phase 5  │
│Session │ │Context │ │Test Covg │ │Conflict  │ │TDD Task  │
│Discover│ │Gather  │ │Analysis  │ │Resolve   │ │Generate  │
│ (ext)  │ │ (ext)  │ │ (local)  │ │(ext,cond)│ │ (local)  │
└────────┘ └────────┘ └──────────┘ └──────────┘ └──────────┘
    ↓           ↓           ↓            ↓            ↓
 sessionId  contextPath  testContext  resolved    IMPL_PLAN.md
            conflict_risk             artifacts   task JSONs

Phase 6: TDD Structure Validation (inline in SKILL.md)

Post-execution verification:
┌──────────────┐     ┌───────────────────┐
│  TDD Verify  │────→│ Coverage Analysis │
│  (local)     │     │ (local)           │
└──────────────┘     └───────────────────┘
phases/03-tdd-        phases/04-tdd-
verify.md             coverage-analysis.md

Key Design Principles

核心设计原则

  1. Pure Orchestrator: Execute phases in sequence, parse outputs, pass context between them
  2. Auto-Continue: All phases run autonomously without user intervention between phases
  3. Subagent Lifecycle: Explicit lifecycle management with spawn_agent → wait → close_agent
  4. Progressive Phase Loading: Phase docs are read on-demand, not all at once
  5. Conditional Execution: Phase 4 only executes when conflict_risk >= medium
  6. TDD-First: Every feature starts with a failing test (Red phase)
  7. Role Path Loading: Subagent roles loaded via path reference in MANDATORY FIRST STEPS
CLI Tool Selection: CLI tool usage is determined semantically from user's task description. Include "use Codex/Gemini/Qwen" in your request for CLI execution.
Task Attachment Model:
  • Skill execute expands workflow by attaching sub-tasks to current TodoWrite
  • When executing a sub-command, its internal tasks are attached to the orchestrator's TodoWrite
  • Orchestrator executes these attached tasks sequentially
  • After completion, attached tasks are collapsed back to high-level phase summary
  • This is task expansion, not external delegation
Auto-Continue Mechanism:
  • TodoList tracks current phase status and dynamically manages task attachment/collapse
  • When each phase finishes executing, automatically execute next pending phase
  • All phases run autonomously without user interaction
  • CONTINUOUS EXECUTION - Do not stop until all phases complete
  1. 纯编排器:按顺序执行各个阶段,解析输出结果,在阶段间传递上下文
  2. 自动续行:所有阶段自主运行,阶段间无需用户干预
  3. 子代理生命周期:通过spawn_agent → wait → close_agent实现明确的生命周期管理
  4. 渐进式阶段加载:按需读取阶段文档,而非一次性加载全部
  5. 条件执行:仅当conflict_risk >= medium时才执行第4阶段
  6. TDD先行:每个功能都从失败的测试开始(红阶段)
  7. 角色路径加载:子代理角色通过MANDATORY FIRST STEPS中的路径引用加载
CLI工具选择:CLI工具的使用由用户任务描述语义决定。若需CLI执行,请在请求中包含"use Codex/Gemini/Qwen"。
任务附加模型:
  • Skill执行时会扩展工作流,将子任务附加到当前TodoWrite
  • 执行子命令时,其内部任务会附加到编排器的TodoWrite
  • 编排器按顺序执行这些附加任务
  • 完成后,附加任务会折叠回高级阶段摘要
  • 此为任务扩展,而非外部委托
自动续行机制:
  • TodoList跟踪当前阶段状态,动态管理任务的附加/折叠
  • 每个阶段执行完成后,自动执行下一个待处理阶段
  • 所有阶段自主运行,无需用户交互
  • 持续执行 - 所有阶段完成前请勿停止

Auto Mode

自动模式

When
--yes
or
-y
: Auto-continue all phases (skip confirmations), use recommended conflict resolutions, skip TDD clarifications.
当使用
--yes
-y
参数时:自动续行所有阶段(跳过确认),使用推荐的冲突解决方案,跳过TDD澄清步骤。

Subagent API Reference

子代理API参考

spawn_agent

spawn_agent

Create a new subagent with task assignment.
javascript
const agentId = spawn_agent({
  message: `
创建一个分配了任务的新子代理。
javascript
const agentId = spawn_agent({
  message: `

TASK ASSIGNMENT

TASK ASSIGNMENT

MANDATORY FIRST STEPS (Agent Execute)

MANDATORY FIRST STEPS (Agent Execute)

  1. Read role definition: ~/.codex/agents/{agent-type}.md (MUST read first)
  2. Read: {projectRoot}/.workflow/project-tech.json
  3. Read: {projectRoot}/.workflow/project-guidelines.json
  1. Read role definition: ~/.codex/agents/{agent-type}.md (MUST read first)
  2. Read: {projectRoot}/.workflow/project-tech.json
  3. Read: {projectRoot}/.workflow/project-guidelines.json

TASK CONTEXT

TASK CONTEXT

${taskContext}
${taskContext}

DELIVERABLES

DELIVERABLES

${deliverables} ` })
undefined
${deliverables} ` })
undefined

wait

wait

Get results from subagent (only way to retrieve results).
javascript
const result = wait({
  ids: [agentId],
  timeout_ms: 600000  // 10 minutes
})

if (result.timed_out) {
  // Handle timeout - can continue waiting or send_input to prompt completion
}
从子代理获取结果(唯一的结果获取方式)。
javascript
const result = wait({
  ids: [agentId],
  timeout_ms: 600000  // 10 minutes
})

if (result.timed_out) {
  // Handle timeout - can continue waiting or send_input to prompt completion
}

send_input

send_input

Continue interaction with active subagent (for clarification or follow-up).
javascript
send_input({
  id: agentId,
  message: `
与活跃子代理继续交互(用于澄清或后续操作)。
javascript
send_input({
  id: agentId,
  message: `

CLARIFICATION ANSWERS

CLARIFICATION ANSWERS

${answers}
${answers}

NEXT STEP

NEXT STEP

Continue with plan generation. ` })
undefined
Continue with plan generation. ` })
undefined

close_agent

close_agent

Clean up subagent resources (irreversible).
javascript
close_agent({ id: agentId })
清理子代理资源(操作不可逆)。
javascript
close_agent({ id: agentId })

Usage

使用方法

workflow-tdd-plan <task description>
workflow-tdd-plan [-y|--yes] "<task description>"
workflow-tdd-plan <task description>
workflow-tdd-plan [-y|--yes] "<task description>"

Flags

标志参数

-y, --yes Skip all confirmations (auto mode)
-y, --yes 跳过所有确认(自动模式)

Arguments

参数

<task description> Task description text, TDD-structured format, or path to .md file
<task description> 任务描述文本、TDD结构化格式内容,或指向.md文件的路径

Examples

示例

workflow-tdd-plan "Build user authentication with tests" # Simple TDD task workflow-tdd-plan "Add JWT auth with email/password and token refresh" # Detailed task workflow-tdd-plan -y "Implement payment processing" # Auto mode workflow-tdd-plan "tdd-requirements.md" # From file
undefined
workflow-tdd-plan "Build user authentication with tests" # 简单TDD任务 workflow-tdd-plan "Add JWT auth with email/password and token refresh" # 详细任务 workflow-tdd-plan -y "Implement payment processing" # 自动模式 workflow-tdd-plan "tdd-requirements.md" # 从文件读取
undefined

TDD Compliance Requirements

TDD合规要求

The Iron Law

铁律

NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST
Enforcement Method:
  • Phase 5:
    implementation_approach
    includes test-first steps (Red → Green → Refactor)
  • Green phase: Includes test-fix-cycle configuration (max 3 iterations)
  • Auto-revert: Triggered when max iterations reached without passing tests
Verification: Phase 6 validates Red-Green-Refactor structure in all generated tasks
NO PRODUCTION CODE WITHOUT A FAILING TEST FIRST
执行方法:
  • 第5阶段:
    implementation_approach
    包含测试先行步骤(红 → 绿 → 重构)
  • 绿阶段:包含测试修复周期配置(最多3次迭代)
  • 自动回滚:当达到最大迭代次数仍未通过测试时触发
验证:第6阶段验证所有生成任务中的红-绿-重构结构

TDD Compliance Checkpoint

TDD合规检查点

CheckpointValidation PhaseEvidence Required
Test-first structurePhase 5
implementation_approach
has 3 steps
Red phase existsPhase 6Step 1:
tdd_phase: "red"
Green phase with test-fixPhase 6Step 2:
tdd_phase: "green"
+ test-fix-cycle
Refactor phase existsPhase 6Step 3:
tdd_phase: "refactor"
检查点验证阶段所需证据
测试先行结构第5阶段
implementation_approach
包含3个步骤
红阶段存在第6阶段步骤1:
tdd_phase: "red"
带测试修复的绿阶段第6阶段步骤2:
tdd_phase: "green"
+ 测试修复周期
重构阶段存在第6阶段步骤3:
tdd_phase: "refactor"

Core TDD Principles

核心TDD原则

Red Flags - STOP and Reassess:
  • Code written before test
  • Test passes immediately (no Red phase witnessed)
  • Cannot explain why test should fail
  • "Just this once" rationalization
  • "Tests after achieve same goals" thinking
Why Order Matters:
  • Tests written after code pass immediately → proves nothing
  • Test-first forces edge case discovery before implementation
  • Tests-after verify what was built, not what's required
危险信号 - 立即停止并重新评估:
  • 先编写代码再编写测试
  • 测试立即通过(未经历红阶段)
  • 无法解释测试为何应该失败
  • “就这一次”的合理化借口
  • “事后测试也能达到同样目标”的想法
顺序为何重要:
  • 代码编写完成后再写的测试会立即通过 → 无法证明任何问题
  • 测试先行迫使在实现前发现边缘情况
  • 事后测试仅验证已构建的内容,而非需求内容

Core Rules

核心规则

  1. Start Immediately: First action is TodoWrite initialization, second action is execute Phase 1
  2. No Preliminary Analysis: Do not read files before Phase 1
  3. Parse Every Output: Extract required data for next phase
  4. Auto-Continue via TodoList: Check TodoList status to execute next pending phase automatically
  5. Track Progress: Update TodoWrite dynamically with task attachment/collapse pattern
  6. TDD Context: All descriptions include "TDD:" prefix
  7. Task Attachment Model: Skill execute attaches sub-tasks to current workflow. Orchestrator executes these attached tasks itself, then collapses them after completion
  8. CRITICAL: DO NOT STOP: Continuous multi-phase workflow. After executing all attached tasks, immediately collapse them and execute next phase
  9. Explicit Lifecycle: Always close_agent after wait completes to free resources
  1. 立即启动:第一个操作是初始化TodoWrite,第二个操作是执行第1阶段
  2. 无预先分析:在第1阶段前请勿读取任何文件
  3. 解析所有输出:提取下一阶段所需的数据
  4. 通过TodoList自动续行:检查TodoList状态以自动执行下一个待处理阶段
  5. 跟踪进度:通过任务附加/折叠模式动态更新TodoWrite
  6. TDD上下文:所有描述均以“TDD:”为前缀
  7. 任务附加模型:Skill执行时会将子任务附加到当前工作流。编排器自行执行这些附加任务,完成后将其折叠
  8. 关键:请勿停止:持续的多阶段工作流。执行完所有附加任务后,立即将其折叠并执行下一阶段
  9. 明确生命周期:wait完成后务必调用close_agent以释放资源

Execution Flow

执行流程

Input Parsing:
   └─ Convert user input to TDD-structured format (TDD:/GOAL/SCOPE/CONTEXT/TEST_FOCUS)

Phase 1: Session Discovery
   └─ Ref: workflow-plan-execute/phases/01-session-discovery.md (external)
      └─ Output: sessionId (WFS-xxx)

Phase 2: Context Gathering
   └─ Ref: workflow-plan-execute/phases/02-context-gathering.md (external)
      ├─ Tasks attached: Analyze structure → Identify integration → Generate package
      └─ Output: contextPath + conflict_risk

Phase 3: Test Coverage Analysis                    ← ATTACHED (3 tasks)
   └─ Ref: phases/01-test-context-gather.md
      ├─ Phase 3.1: Detect test framework
      ├─ Phase 3.2: Analyze existing test coverage
      └─ Phase 3.3: Identify coverage gaps
      └─ Output: test-context-package.json           ← COLLAPSED

Phase 4: Conflict Resolution (conditional)
   └─ Decision (conflict_risk check):
      ├─ conflict_risk ≥ medium → Inline conflict resolution (within Phase 2)
      │   ├─ Tasks attached: Detect conflicts → Present to user → Apply strategies
      │   └─ Output: Modified brainstorm artifacts     ← COLLAPSED
      └─ conflict_risk < medium → Skip to Phase 5

Phase 5: TDD Task Generation                       ← ATTACHED (3 tasks)
   └─ Ref: phases/02-task-generate-tdd.md
      ├─ Phase 5.1: Discovery - analyze TDD requirements
      ├─ Phase 5.2: Planning - design Red-Green-Refactor cycles
      └─ Phase 5.3: Output - generate IMPL tasks with internal TDD phases
      └─ Output: IMPL-*.json, IMPL_PLAN.md           ← COLLAPSED

Phase 6: TDD Structure Validation (inline)
   └─ Internal validation + summary returned
   └─ Recommend: plan-verify (external)

Return:
   └─ Summary with recommended next steps
输入解析:
   └─ 将用户输入转换为TDD结构化格式(TDD:/GOAL/SCOPE/CONTEXT/TEST_FOCUS)

第1阶段:会话发现
   └─ 参考:workflow-plan-execute/phases/01-session-discovery.md(外部)
      └─ 输出:sessionId(WFS-xxx)

第2阶段:上下文收集
   └─ 参考:workflow-plan-execute/phases/02-context-gathering.md(外部)
      ├─ 附加任务:分析结构 → 识别集成点 → 生成包
      └─ 输出:contextPath + conflict_risk

第3阶段:测试覆盖率分析                    ← 已附加(3个任务)
   └─ 参考:phases/01-test-context-gather.md
      ├─ 第3.1阶段:检测测试框架
      ├─ 第3.2阶段:分析现有测试覆盖率
      └─ 第3.3阶段:识别覆盖率缺口
      └─ 输出:test-context-package.json           ← 已折叠

第4阶段:冲突解决(条件执行)
   └─ 决策(检查conflict_risk):
      ├─ conflict_risk ≥ medium → 内联冲突解决(在第2阶段内)
      │   ├─ 附加任务:检测冲突 → 呈现给用户 → 应用解决方案
      │   └─ 输出:修改后的头脑风暴产物     ← 已折叠
      └─ conflict_risk < medium → 跳至第5阶段

第5阶段:TDD任务生成                       ← 已附加(3个任务)
   └─ 参考:phases/02-task-generate-tdd.md
      ├─ 第5.1阶段:发现 - 分析TDD需求
      ├─ 第5.2阶段:规划 - 设计红-绿-重构周期
      └─ 第5.3阶段:输出 - 生成包含内部TDD阶段的IMPL任务
      └─ 输出:IMPL-*.json, IMPL_PLAN.md           ← 已折叠

第6阶段:TDD结构验证(内联)
   └─ 执行内部验证并返回摘要
   └─ 推荐:plan-verify(外部)

返回:
   └─ 包含推荐后续步骤的摘要

Phase Reference Documents

阶段参考文档

Local phases (read on-demand when phase executes):
PhaseDocumentPurpose
Phase 3phases/01-test-context-gather.mdTest coverage context gathering via test-context-search-agent
Phase 5phases/02-task-generate-tdd.mdTDD task JSON generation via action-planning-agent
External phases (from workflow-plan-execute skill):
PhaseDocumentPurpose
Phase 1workflow-plan-execute/phases/01-session-discovery.mdSession creation/discovery
Phase 2workflow-plan-execute/phases/02-context-gathering.mdProject context collection + inline conflict resolution
Post-execution verification:
PhaseDocumentPurpose
TDD Verifyphases/03-tdd-verify.mdTDD compliance verification with quality gate
Coverage Analysisphases/04-tdd-coverage-analysis.mdTest coverage and cycle analysis (called by TDD Verify)
本地阶段(阶段执行时按需读取):
阶段文档用途
第3阶段phases/01-test-context-gather.md通过test-context-search-agent收集测试覆盖率上下文
第5阶段phases/02-task-generate-tdd.md通过action-planning-agent生成TDD任务JSON
外部阶段(来自workflow-plan-execute skill):
阶段文档用途
第1阶段workflow-plan-execute/phases/01-session-discovery.md会话创建/发现
第2阶段workflow-plan-execute/phases/02-context-gathering.md项目上下文收集 + 内联冲突解决
执行后验证:
阶段文档用途
TDD验证phases/03-tdd-verify.md带有质量门的TDD合规验证
覆盖率分析phases/04-tdd-coverage-analysis.md测试覆盖率和周期分析(由TDD验证调用)

6-Phase Execution

6阶段执行

Phase 1: Session Discovery

第1阶段:会话发现

Step 1.1: Execute - Session discovery and initialization
Read and execute:
workflow-plan-execute/phases/01-session-discovery.md
with
--type tdd --auto "TDD: [structured-description]"
TDD Structured Format:
TDD: [Feature Name]
GOAL: [Objective]
SCOPE: [Included/excluded]
CONTEXT: [Background]
TEST_FOCUS: [Test scenarios]
Parse: Extract sessionId
TodoWrite: Mark phase 1 completed, phase 2 in_progress
After Phase 1: Return to user showing Phase 1 results, then auto-continue to Phase 2

步骤1.1:执行 - 会话发现与初始化
读取并执行:
workflow-plan-execute/phases/01-session-discovery.md
,参数为
--type tdd --auto "TDD: [structured-description]"
TDD结构化格式:
TDD: [Feature Name]
GOAL: [Objective]
SCOPE: [Included/excluded]
CONTEXT: [Background]
TEST_FOCUS: [Test scenarios]
解析:提取sessionId
TodoWrite:标记第1阶段已完成,第2阶段进行中
第1阶段后:向用户返回第1阶段结果,然后自动续行至第2阶段

Phase 2: Context Gathering

第2阶段:上下文收集

Step 2.1: Execute - Context gathering and analysis
Read and execute:
workflow-plan-execute/phases/02-context-gathering.md
with
--session [sessionId] "TDD: [structured-description]"
Use Same Structured Description: Pass the same structured format from Phase 1
Input:
sessionId
from Phase 1
Parse Output:
  • Extract: context-package.json path (store as
    contextPath
    )
  • Typical pattern:
    {projectRoot}/.workflow/active/[sessionId]/.process/context-package.json
Validation:
  • Context package path extracted
  • File exists and is valid JSON
TodoWrite: Mark phase 2 completed, phase 3 in_progress
After Phase 2: Return to user showing Phase 2 results, then auto-continue to Phase 3

步骤2.1:执行 - 上下文收集与分析
读取并执行:
workflow-plan-execute/phases/02-context-gathering.md
,参数为
--session [sessionId] "TDD: [structured-description]"
使用相同的结构化描述:传递与第1阶段相同的结构化格式内容
输入:来自第1阶段的
sessionId
解析输出:
  • 提取:context-package.json路径(存储为
    contextPath
  • 典型路径:
    {projectRoot}/.workflow/active/[sessionId]/.process/context-package.json
验证:
  • 已提取上下文包路径
  • 文件存在且为有效JSON
TodoWrite:标记第2阶段已完成,第3阶段进行中
第2阶段后:向用户返回第2阶段结果,然后自动续行至第3阶段

Phase 3: Test Coverage Analysis

第3阶段:测试覆盖率分析

Step 3.1: Execute - Test coverage analysis and framework detection
Read and execute:
phases/01-test-context-gather.md
with
--session [sessionId]
Purpose: Analyze existing codebase for:
  • Existing test patterns and conventions
  • Current test coverage
  • Related components and integration points
  • Test framework detection
Parse: Extract testContextPath (
{projectRoot}/.workflow/active/[sessionId]/.process/test-context-package.json
)
TodoWrite Update (Phase 3 - tasks attached):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "in_progress", "activeForm": "Executing test coverage analysis"},
  {"content": "  → Detect test framework and conventions", "status": "in_progress", "activeForm": "Detecting test framework"},
  {"content": "  → Analyze existing test coverage", "status": "pending", "activeForm": "Analyzing test coverage"},
  {"content": "  → Identify coverage gaps", "status": "pending", "activeForm": "Identifying coverage gaps"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
Note: Skill execute attaches test-context-gather's 3 tasks. Orchestrator executes these tasks.
Next Action: Tasks attached → Execute Phase 3.1-3.3 sequentially
TodoWrite Update (Phase 3 completed - tasks collapsed):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
After Phase 3: Return to user showing test coverage results, then auto-continue to Phase 4/5

步骤3.1:执行 - 测试覆盖率分析与框架检测
读取并执行:
phases/01-test-context-gather.md
,参数为
--session [sessionId]
目的:分析现有代码库的以下内容:
  • 现有测试模式与约定
  • 当前测试覆盖率
  • 相关组件与集成点
  • 测试框架检测
解析:提取testContextPath(
{projectRoot}/.workflow/active/[sessionId]/.process/test-context-package.json
TodoWrite更新(第3阶段 - 任务已附加):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "in_progress", "activeForm": "Executing test coverage analysis"},
  {"content": "  → Detect test framework and conventions", "status": "in_progress", "activeForm": "Detecting test framework"},
  {"content": "  → Analyze existing test coverage", "status": "pending", "activeForm": "Analyzing test coverage"},
  {"content": "  → Identify coverage gaps", "status": "pending", "activeForm": "Identifying coverage gaps"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
注意:Skill执行时会附加test-context-gather的3个任务。编排器执行这些任务。
下一步操作:任务已附加 → 依次执行第3.1-3.3阶段
TodoWrite更新(第3阶段完成 - 任务已折叠):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
第3阶段后:向用户返回测试覆盖率结果,然后自动续行至第4/5阶段

Phase 4: Conflict Resolution (Optional)

第4阶段:冲突解决(可选)

Trigger: Only execute when context-package.json indicates conflict_risk is "medium" or "high"
Step 4.1: Execute - Conflict detection and resolution
Conflict resolution is handled inline within Phase 2 (context-gathering). When conflict_risk >= medium, Phase 2 automatically performs detection and resolution.
Input:
  • sessionId from Phase 1
  • contextPath from Phase 2
  • conflict_risk from context-package.json
Parse Output:
  • Extract: Execution status (success/skipped/failed)
  • Verify: conflict-resolution.json file path (if executed)
Skip Behavior:
  • If conflict_risk is "none" or "low", skip directly to Phase 5
  • Display: "No significant conflicts detected, proceeding to TDD task generation"
TodoWrite Update (Phase 4 - tasks attached, if conflict_risk >= medium):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 4: Conflict Resolution", "status": "in_progress", "activeForm": "Executing conflict resolution"},
  {"content": "  → Detect conflicts with CLI analysis", "status": "in_progress", "activeForm": "Detecting conflicts"},
  {"content": "  → Log and analyze detected conflicts", "status": "pending", "activeForm": "Analyzing conflicts"},
  {"content": "  → Apply resolution strategies", "status": "pending", "activeForm": "Applying resolution strategies"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
TodoWrite Update (Phase 4 completed - tasks collapsed):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 4: Conflict Resolution", "status": "completed", "activeForm": "Executing conflict resolution"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
After Phase 4: Return to user showing conflict resolution results, then auto-continue to Phase 5
Memory State Check:
  • Evaluate current context window usage and memory state
  • If memory usage is high (>110K tokens or approaching context limits):
    Step 4.5: Execute - Memory compaction (external skill: compact)
    • This optimizes memory before proceeding to Phase 5
  • Memory compaction is particularly important after analysis phase which may generate extensive documentation

触发条件:仅当context-package.json中conflict_risk为"medium"或"high"时执行
步骤4.1:执行 - 冲突检测与解决
冲突解决在第2阶段(上下文收集)内联处理。当conflict_risk >= medium时,第2阶段自动执行检测与解决。
输入:
  • 来自第1阶段的sessionId
  • 来自第2阶段的contextPath
  • 来自context-package.json的conflict_risk
解析输出:
  • 提取:执行状态(成功/已跳过/失败)
  • 验证:conflict-resolution.json文件路径(若已执行)
跳过行为:
  • 若conflict_risk为"none"或"low",直接跳至第5阶段
  • 显示:"未检测到重大冲突,继续执行TDD任务生成"
TodoWrite更新(第4阶段 - 任务已附加,若conflict_risk >= medium):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 4: Conflict Resolution", "status": "in_progress", "activeForm": "Executing conflict resolution"},
  {"content": "  → Detect conflicts with CLI analysis", "status": "in_progress", "activeForm": "Detecting conflicts"},
  {"content": "  → Log and analyze detected conflicts", "status": "pending", "activeForm": "Analyzing conflicts"},
  {"content": "  → Apply resolution strategies", "status": "pending", "activeForm": "Applying resolution strategies"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
TodoWrite更新(第4阶段完成 - 任务已折叠):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 4: Conflict Resolution", "status": "completed", "activeForm": "Executing conflict resolution"},
  {"content": "Phase 5: TDD Task Generation", "status": "pending", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
第4阶段后:向用户返回冲突解决结果,然后自动续行至第5阶段
内存状态检查:
  • 评估当前上下文窗口使用情况与内存状态
  • 若内存使用率过高(>110K tokens或接近上下文限制):
    步骤4.5:执行 - 内存压缩(外部skill: compact)
    • 此操作会在进入第5阶段前优化内存
  • 分析阶段可能生成大量文档,因此内存压缩尤为重要

Phase 5: TDD Task Generation

第5阶段:TDD任务生成

Step 5.1: Execute - TDD task generation via action-planning-agent with Phase 0 user configuration
Read and execute:
phases/02-task-generate-tdd.md
with
--session [sessionId]
Note: Phase 0 now includes:
  • Supplementary materials collection (file paths or inline content)
  • Execution method preference (Agent/Hybrid/CLI)
  • CLI tool preference (Codex/Gemini/Qwen/Auto)
  • These preferences are passed to agent for task generation
Parse: Extract feature count, task count, CLI execution IDs assigned
Validate:
  • IMPL_PLAN.md exists (unified plan with TDD Implementation Tasks section)
  • IMPL-*.json files exist (one per feature, or container + subtasks for complex features)
  • TODO_LIST.md exists with internal TDD phase indicators
  • Each IMPL task includes:
    • meta.tdd_workflow: true
    • meta.cli_execution_id: {session_id}-{task_id}
    • meta.cli_execution: { "strategy": "new|resume|fork|merge_fork", ... }
    • flow_control.implementation_approach
      with exactly 3 steps (red/green/refactor)
    • Green phase includes test-fix-cycle configuration
    • context.focus_paths
      : absolute or clear relative paths
    • flow_control.pre_analysis
      : includes exploration integration_points analysis
  • IMPL_PLAN.md contains workflow_type: "tdd" in frontmatter
  • Task count <=18 (compliance with hard limit)
Red Flag Detection (Non-Blocking Warnings):
  • Task count >18:
    Warning: Task count exceeds hard limit - request re-scope
  • Missing cli_execution_id:
    Warning: Task lacks CLI execution ID for resume support
  • Missing test-fix-cycle:
    Warning: Green phase lacks auto-revert configuration
  • Generic task names:
    Warning: Vague task names suggest unclear TDD cycles
  • Missing focus_paths:
    Warning: Task lacks clear file scope for implementation
Action: Log warnings to
{projectRoot}/.workflow/active/[sessionId]/.process/tdd-warnings.log
(non-blocking)
TodoWrite Update (Phase 5 - tasks attached):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 5: TDD Task Generation", "status": "in_progress", "activeForm": "Executing TDD task generation"},
  {"content": "  → Discovery - analyze TDD requirements", "status": "in_progress", "activeForm": "Analyzing TDD requirements"},
  {"content": "  → Planning - design Red-Green-Refactor cycles", "status": "pending", "activeForm": "Designing TDD cycles"},
  {"content": "  → Output - generate IMPL tasks with internal TDD phases", "status": "pending", "activeForm": "Generating TDD tasks"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
TodoWrite Update (Phase 5 completed - tasks collapsed):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 5: TDD Task Generation", "status": "completed", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "in_progress", "activeForm": "Validating TDD structure"}
]
步骤5.1:执行 - 通过action-planning-agent结合第0阶段用户配置生成TDD任务
读取并执行:
phases/02-task-generate-tdd.md
,参数为
--session [sessionId]
注意:第0阶段现在包含:
  • 补充材料收集(文件路径或内联内容)
  • 执行方式偏好(Agent/Hybrid/CLI)
  • CLI工具偏好(Codex/Gemini/Qwen/Auto)
  • 这些偏好会传递给代理用于任务生成
解析:提取功能数量、任务数量、已分配的CLI执行ID
验证:
  • IMPL_PLAN.md存在(包含TDD实现任务章节的统一计划)
  • IMPL-*.json文件存在(每个功能对应一个,复杂功能包含容器+子任务)
  • TODO_LIST.md存在且包含内部TDD阶段标识
  • 每个IMPL任务包含:
    • meta.tdd_workflow: true
    • meta.cli_execution_id: {session_id}-{task_id}
    • meta.cli_execution: { "strategy": "new|resume|fork|merge_fork", ... }
    • flow_control.implementation_approach
      包含恰好3个步骤(红/绿/重构)
    • 绿阶段包含测试修复周期配置
    • context.focus_paths
      : 绝对路径或清晰的相对路径
    • flow_control.pre_analysis
      : 包含探索性集成点分析
  • IMPL_PLAN.md的前置元数据中包含workflow_type: "tdd"
  • 任务数量<=18(符合硬限制)
危险信号检测(非阻塞警告):
  • 任务数量>18:
    Warning: Task count exceeds hard limit - request re-scope
  • 缺少cli_execution_id:
    Warning: Task lacks CLI execution ID for resume support
  • 缺少测试修复周期:
    Warning: Green phase lacks auto-revert configuration
  • 通用任务名称:
    Warning: Vague task names suggest unclear TDD cycles
  • 缺少focus_paths:
    Warning: Task lacks clear file scope for implementation
操作:将警告日志记录到
{projectRoot}/.workflow/active/[sessionId]/.process/tdd-warnings.log
(非阻塞)
TodoWrite更新(第5阶段 - 任务已附加):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 5: TDD Task Generation", "status": "in_progress", "activeForm": "Executing TDD task generation"},
  {"content": "  → Discovery - analyze TDD requirements", "status": "in_progress", "activeForm": "Analyzing TDD requirements"},
  {"content": "  → Planning - design Red-Green-Refactor cycles", "status": "pending", "activeForm": "Designing TDD cycles"},
  {"content": "  → Output - generate IMPL tasks with internal TDD phases", "status": "pending", "activeForm": "Generating TDD tasks"},
  {"content": "Phase 6: TDD Structure Validation", "status": "pending", "activeForm": "Validating TDD structure"}
]
TodoWrite更新(第5阶段完成 - 任务已折叠):
json
[
  {"content": "Phase 1: Session Discovery", "status": "completed", "activeForm": "Executing session discovery"},
  {"content": "Phase 2: Context Gathering", "status": "completed", "activeForm": "Executing context gathering"},
  {"content": "Phase 3: Test Coverage Analysis", "status": "completed", "activeForm": "Executing test coverage analysis"},
  {"content": "Phase 5: TDD Task Generation", "status": "completed", "activeForm": "Executing TDD task generation"},
  {"content": "Phase 6: TDD Structure Validation", "status": "in_progress", "activeForm": "Validating TDD structure"}
]

Phase 6: TDD Structure Validation & Action Plan Verification (RECOMMENDED)

第6阶段:TDD结构验证与行动计划验证(推荐)

Internal validation first, then recommend external verification
Internal Validation:
  1. Each task contains complete TDD workflow (Red-Green-Refactor internally)
  2. Task structure validation:
    • meta.tdd_workflow: true
      in all IMPL tasks
    • meta.cli_execution_id
      present (format: {session_id}-{task_id})
    • meta.cli_execution
      strategy assigned (new/resume/fork/merge_fork)
    • flow_control.implementation_approach
      has exactly 3 steps
    • Each step has correct
      tdd_phase
      : "red", "green", "refactor"
    • context.focus_paths
      are absolute or clear relative paths
    • flow_control.pre_analysis
      includes exploration integration analysis
  3. Dependency validation:
    • Sequential features: IMPL-N depends_on ["IMPL-(N-1)"] if needed
    • Complex features: IMPL-N.M depends_on ["IMPL-N.(M-1)"] for subtasks
    • CLI execution strategies correctly assigned based on dependency graph
  4. Agent assignment: All IMPL tasks use @code-developer
  5. Test-fix cycle: Green phase step includes test-fix-cycle logic with max_iterations
  6. Task count: Total tasks <=18 (simple + subtasks hard limit)
  7. User configuration:
    • Execution method choice reflected in task structure
    • CLI tool preference documented in implementation guidance (if CLI selected)
Red Flag Checklist (from TDD best practices):
  • No tasks skip Red phase (
    tdd_phase: "red"
    exists in step 1)
  • Test files referenced in Red phase (explicit paths, not placeholders)
  • Green phase has test-fix-cycle with
    max_iterations
    configured
  • Refactor phase has clear completion criteria
Non-Compliance Warning Format:
Warning TDD Red Flag: [issue description]
   Task: [IMPL-N]
   Recommendation: [action to fix]
Evidence Gathering (Before Completion Claims):
bash
undefined
先执行内部验证,再推荐外部验证
内部验证:
  1. 每个任务包含完整的TDD工作流(内部红-绿-重构)
  2. 任务结构验证:
    • 所有IMPL任务中
      meta.tdd_workflow: true
    • 存在
      meta.cli_execution_id
      (格式: {session_id}-{task_id})
    • 已分配
      meta.cli_execution
      策略(new/resume/fork/merge_fork)
    • flow_control.implementation_approach
      包含恰好3个步骤
    • 每个步骤的
      tdd_phase
      正确:"red"、"green"、"refactor"
    • context.focus_paths
      为绝对路径或清晰的相对路径
    • flow_control.pre_analysis
      包含探索性集成分析
  3. 依赖验证:
    • 顺序功能:若需要,IMPL-N depends_on ["IMPL-(N-1)"]
    • 复杂功能:IMPL-N.M depends_on ["IMPL-N.(M-1)"](子任务)
    • 根据依赖图正确分配CLI执行策略
  4. 代理分配:所有IMPL任务使用@code-developer
  5. 测试修复周期:绿阶段步骤包含带有max_iterations的测试修复周期逻辑
  6. 任务数量:总任务数<=18(简单任务+子任务硬限制)
  7. 用户配置:
    • 任务结构反映执行方式选择
    • 若选择CLI,实现指南中记录了CLI工具偏好
危险信号检查表(来自TDD最佳实践):
  • 无任务跳过红阶段(步骤1中存在
    tdd_phase: "red"
  • 红阶段引用了测试文件(明确路径,而非占位符)
  • 绿阶段带有配置了
    max_iterations
    的测试修复周期
  • 重构阶段有清晰的完成标准
不合规警告格式:
Warning TDD Red Flag: [issue description]
   Task: [IMPL-N]
   Recommendation: [action to fix]
证据收集(完成声明前):
bash
undefined

Verify session artifacts exist

验证会话产物存在

ls -la ${projectRoot}/.workflow/active/[sessionId]/{IMPL_PLAN.md,TODO_LIST.md} ls -la ${projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json
ls -la ${projectRoot}/.workflow/active/[sessionId]/{IMPL_PLAN.md,TODO_LIST.md} ls -la ${projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json

Count generated artifacts

统计生成的产物数量

echo "IMPL tasks: $(ls ${projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json 2>/dev/null | wc -l)"
echo "IMPL tasks: $(ls ${projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json 2>/dev/null | wc -l)"

Sample task structure verification (first task)

验证任务结构示例(第一个任务)

jq '{id, tdd: .meta.tdd_workflow, cli_id: .meta.cli_execution_id, phases: [.flow_control.implementation_approach[].tdd_phase]}'
"$(ls ${projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json | head -1)"

**Evidence Required Before Summary**:
| Evidence Type | Verification Method | Pass Criteria |
|---------------|---------------------|---------------|
| File existence | `ls -la` artifacts | All files present |
| Task count | Count IMPL-*.json | Count matches claims (<=18) |
| TDD structure | jq sample extraction | Shows red/green/refactor + cli_execution_id |
| CLI execution IDs | jq extraction | All tasks have cli_execution_id assigned |
| Warning log | Check tdd-warnings.log | Logged (may be empty) |

**Return Summary**:
TDD Planning complete for session: [sessionId]
Features analyzed: [N] Total tasks: [M] (1 task per simple feature + subtasks for complex features)
Task breakdown:
  • Simple features: [K] tasks (IMPL-1 to IMPL-K)
  • Complex features: [L] features with [P] subtasks
  • Total task count: [M] (within 18-task hard limit)
Structure:
  • IMPL-1: {Feature 1 Name} (Internal: Red → Green → Refactor)
  • IMPL-2: {Feature 2 Name} (Internal: Red → Green → Refactor)
  • IMPL-3: {Complex Feature} (Container)
    • IMPL-3.1: {Sub-feature A} (Internal: Red → Green → Refactor)
    • IMPL-3.2: {Sub-feature B} (Internal: Red → Green → Refactor) [...]
Plans generated:
  • Unified Implementation Plan: {projectRoot}/.workflow/active/[sessionId]/IMPL_PLAN.md (includes TDD Implementation Tasks section with workflow_type: "tdd")
  • Task List: {projectRoot}/.workflow/active/[sessionId]/TODO_LIST.md (with internal TDD phase indicators and CLI execution strategies)
  • Task JSONs: {projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json (with cli_execution_id and execution strategies for resume support)
TDD Configuration:
  • Each task contains complete Red-Green-Refactor cycle
  • Green phase includes test-fix cycle (max 3 iterations)
  • Auto-revert on max iterations reached
  • CLI execution strategies: new/resume/fork/merge_fork based on dependency graph
User Configuration Applied:
  • Execution Method: [agent|hybrid|cli]
  • CLI Tool Preference: [codex|gemini|qwen|auto]
  • Supplementary Materials: [included|none]
  • Task generation follows cli-tools-usage.md guidelines
ACTION REQUIRED: Before execution, ensure you understand WHY each Red phase test is expected to fail. This is crucial for valid TDD - if you don't know why the test fails, you can't verify it tests the right thing.
Recommended Next Steps:
  1. plan-verify (external) --session [sessionId] # Verify TDD plan quality and dependencies
  2. workflow:execute (external) --session [sessionId] # Start TDD execution with CLI strategies
  3. phases/03-tdd-verify.md [sessionId] # Post-execution TDD compliance check
Quality Gate: Consider running plan-verify to validate TDD task structure, dependencies, and CLI execution strategies
undefined
jq '{id, tdd: .meta.tdd_workflow, cli_id: .meta.cli_execution_id, phases: [.flow_control.implementation_approach[].tdd_phase]}'
"$(ls ${projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json | head -1)"

**摘要前所需证据**:
| 证据类型 | 验证方法 | 通过标准 |
|---------------|---------------------|---------------|
| 文件存在性 | `ls -la` 产物 | 所有文件均存在 |
| 任务数量 | 统计IMPL-*.json数量 | 数量与声明一致(<=18) |
| TDD结构 | jq提取示例 | 显示红/绿/重构 + cli_execution_id |
| CLI执行ID | jq提取 | 所有任务均有cli_execution_id |
| 警告日志 | 检查tdd-warnings.log | 已记录(可能为空) |

**返回摘要**:
TDD Planning complete for session: [sessionId]
Features analyzed: [N] Total tasks: [M] (1 task per simple feature + subtasks for complex features)
Task breakdown:
  • Simple features: [K] tasks (IMPL-1 to IMPL-K)
  • Complex features: [L] features with [P] subtasks
  • Total task count: [M] (within 18-task hard limit)
Structure:
  • IMPL-1: {Feature 1 Name} (Internal: Red → Green → Refactor)
  • IMPL-2: {Feature 2 Name} (Internal: Red → Green → Refactor)
  • IMPL-3: {Complex Feature} (Container)
    • IMPL-3.1: {Sub-feature A} (Internal: Red → Green → Refactor)
    • IMPL-3.2: {Sub-feature B} (Internal: Red → Green → Refactor) [...]
Plans generated:
  • Unified Implementation Plan: {projectRoot}/.workflow/active/[sessionId]/IMPL_PLAN.md (includes TDD Implementation Tasks section with workflow_type: "tdd")
  • Task List: {projectRoot}/.workflow/active/[sessionId]/TODO_LIST.md (with internal TDD phase indicators and CLI execution strategies)
  • Task JSONs: {projectRoot}/.workflow/active/[sessionId]/.task/IMPL-*.json (with cli_execution_id and execution strategies for resume support)
TDD Configuration:
  • Each task contains complete Red-Green-Refactor cycle
  • Green phase includes test-fix cycle (max 3 iterations)
  • Auto-revert on max iterations reached
  • CLI execution strategies: new/resume/fork/merge_fork based on dependency graph
User Configuration Applied:
  • Execution Method: [agent|hybrid|cli]
  • CLI Tool Preference: [codex|gemini|qwen|auto]
  • Supplementary Materials: [included|none]
  • Task generation follows cli-tools-usage.md guidelines
ACTION REQUIRED: Before execution, ensure you understand WHY each Red phase test is expected to fail. This is crucial for valid TDD - if you don't know why the test fails, you can't verify it tests the right thing.
Recommended Next Steps:
  1. plan-verify (external) --session [sessionId] # Verify TDD plan quality and dependencies
  2. workflow:execute (external) --session [sessionId] # Start TDD execution with CLI strategies
  3. phases/03-tdd-verify.md [sessionId] # Post-execution TDD compliance check
Quality Gate: Consider running plan-verify to validate TDD task structure, dependencies, and CLI execution strategies
undefined

Input Processing

输入处理

Convert user input to TDD-structured format:
Simple text → Add TDD context Detailed text → Extract components with TEST_FOCUS File/Issue → Read and structure with TDD
将用户输入转换为TDD结构化格式:
简单文本 → 添加TDD上下文 详细文本 → 提取包含TEST_FOCUS的组件 文件/问题 → 读取并按TDD格式结构化

Data Flow

数据流

User Input (task description)
[Convert to TDD Structured Format]
    ↓ TDD Structured Description:
    ↓   TDD: [Feature Name]
    ↓   GOAL: [objective]
    ↓   SCOPE: [boundaries]
    ↓   CONTEXT: [background]
    ↓   TEST_FOCUS: [test scenarios]
Phase 1: session:start --type tdd --auto "TDD: structured-description"
    ↓ Output: sessionId
Phase 2: context-gather --session sessionId "TDD: structured-description"
    ↓ Output: contextPath + conflict_risk
Phase 3: test-context-gather --session sessionId
    ↓ Output: testContextPath (test-context-package.json)
Phase 4: conflict-resolution [AUTO-TRIGGERED if conflict_risk >= medium]
    ↓ Output: Modified brainstorm artifacts
    ↓ Skip if conflict_risk is none/low → proceed directly to Phase 5
Phase 5: task-generate-tdd --session sessionId
    ↓ Output: IMPL_PLAN.md, task JSONs, TODO_LIST.md
Phase 6: Internal validation + summary
Return summary to user
User Input (task description)
[Convert to TDD Structured Format]
    ↓ TDD Structured Description:
    ↓   TDD: [Feature Name]
    ↓   GOAL: [objective]
    ↓   SCOPE: [boundaries]
    ↓   CONTEXT: [background]
    ↓   TEST_FOCUS: [test scenarios]
Phase 1: session:start --type tdd --auto "TDD: structured-description"
    ↓ Output: sessionId
Phase 2: context-gather --session sessionId "TDD: structured-description"
    ↓ Output: contextPath + conflict_risk
Phase 3: test-context-gather --session sessionId
    ↓ Output: testContextPath (test-context-package.json)
Phase 4: conflict-resolution [AUTO-TRIGGERED if conflict_risk >= medium]
    ↓ Output: Modified brainstorm artifacts
    ↓ Skip if conflict_risk is none/low → proceed directly to Phase 5
Phase 5: task-generate-tdd --session sessionId
    ↓ Output: IMPL_PLAN.md, task JSONs, TODO_LIST.md
Phase 6: Internal validation + summary
Return summary to user

TodoWrite Pattern

TodoWrite模式

Core Concept: Dynamic task attachment and collapse for TDD workflow with test coverage analysis and Red-Green-Refactor cycle generation.
核心概念:针对TDD工作流的动态任务附加与折叠,包含测试覆盖率分析及红-绿-重构周期生成。

Key Principles

关键原则

  1. Task Attachment (when Skill executed):
    • Sub-command's internal tasks are attached to orchestrator's TodoWrite
    • First attached task marked as
      in_progress
      , others as
      pending
    • Orchestrator executes these attached tasks sequentially
  2. Task Collapse (after sub-tasks complete):
    • Remove detailed sub-tasks from TodoWrite
    • Collapse to high-level phase summary
    • Maintains clean orchestrator-level view
  3. Continuous Execution:
    • After collapse, automatically proceed to next pending phase
    • No user intervention required between phases
    • TodoWrite dynamically reflects current execution state
Lifecycle Summary: Initial pending tasks → Phase executed (tasks ATTACHED) → Sub-tasks executed sequentially → Phase completed (tasks COLLAPSED to summary) → Next phase begins (conditional Phase 4 if conflict_risk >= medium) → Repeat until all phases complete.
  1. 任务附加(Skill执行时):
    • 子命令的内部任务附加到编排器的TodoWrite
    • 第一个附加任务标记为
      in_progress
      ,其余为
      pending
    • 编排器按顺序执行这些附加任务
  2. 任务折叠(子任务完成后):
    • 从TodoWrite中移除详细子任务
    • 折叠为高级阶段摘要
    • 保持编排器层面的清晰视图
  3. 持续执行:
    • 折叠后,自动进入下一个待处理阶段
    • 阶段间无需用户干预
    • TodoWrite动态反映当前执行状态
生命周期摘要:初始待处理任务 → 执行阶段(任务已附加) → 按顺序执行子任务 → 阶段完成(任务折叠为摘要) → 进入下一阶段(若conflict_risk >= medium则执行条件性第4阶段) → 重复直至所有阶段完成。

TDD-Specific Features

TDD特定功能

  • Phase 3: Test coverage analysis detects existing patterns and gaps
  • Phase 5: Generated IMPL tasks contain internal Red-Green-Refactor cycles
  • Conditional Phase 4: Conflict resolution only if conflict_risk >= medium
Note: See individual Phase descriptions (Phase 3, 4, 5) for detailed TodoWrite Update examples with full JSON structures.
  • 第3阶段:测试覆盖率分析检测现有模式与缺口
  • 第5阶段:生成的IMPL任务包含内部红-绿-重构周期
  • 条件性第4阶段:仅当conflict_risk >= medium时执行冲突解决
注意:请查看各个阶段描述(第3、4、5阶段)获取带有完整JSON结构的详细TodoWrite更新示例。

Execution Flow Diagram

执行流程图

TDD Workflow Orchestrator
├─ Phase 1: Session Discovery
│  └─ workflow-plan-execute/phases/01-session-discovery.md --auto
│     └─ Returns: sessionId
├─ Phase 2: Context Gathering
│  └─ workflow-plan-execute/phases/02-context-gathering.md
│     └─ Returns: context-package.json path
├─ Phase 3: Test Coverage Analysis                    ← ATTACHED (3 tasks)
│  └─ phases/01-test-context-gather.md
│     ├─ Phase 3.1: Detect test framework
│     ├─ Phase 3.2: Analyze existing test coverage
│     └─ Phase 3.3: Identify coverage gaps
│     └─ Returns: test-context-package.json           ← COLLAPSED
├─ Phase 4: Conflict Resolution (conditional)
│  IF conflict_risk >= medium:
│  └─ Inline within Phase 2 context-gathering          ← ATTACHED (3 tasks)
│     ├─ Phase 4.1: Detect conflicts with CLI
│     ├─ Phase 4.2: Log and analyze detected conflicts
│     └─ Phase 4.3: Apply resolution strategies
│     └─ Returns: conflict-resolution.json            ← COLLAPSED
│  ELSE:
│  └─ Skip to Phase 5
├─ Phase 5: TDD Task Generation                       ← ATTACHED (3 tasks)
│  └─ phases/02-task-generate-tdd.md
│     ├─ Phase 5.1: Discovery - analyze TDD requirements
│     ├─ Phase 5.2: Planning - design Red-Green-Refactor cycles
│     └─ Phase 5.3: Output - generate IMPL tasks with internal TDD phases
│     └─ Returns: IMPL-*.json, IMPL_PLAN.md           ← COLLAPSED
│        (Each IMPL task contains internal Red-Green-Refactor cycle)
└─ Phase 6: TDD Structure Validation
   └─ Internal validation + summary returned
   └─ Recommend: plan-verify (external)

Key Points:
• ← ATTACHED: Sub-tasks attached to orchestrator TodoWrite
• ← COLLAPSED: Sub-tasks executed and collapsed to phase summary
• TDD-specific: Each generated IMPL task contains complete Red-Green-Refactor cycle
TDD Workflow Orchestrator
├─ Phase 1: Session Discovery
│  └─ workflow-plan-execute/phases/01-session-discovery.md --auto
│     └─ Returns: sessionId
├─ Phase 2: Context Gathering
│  └─ workflow-plan-execute/phases/02-context-gathering.md
│     └─ Returns: context-package.json path
├─ Phase 3: Test Coverage Analysis                    ← ATTACHED (3 tasks)
│  └─ phases/01-test-context-gather.md
│     ├─ Phase 3.1: Detect test framework
│     ├─ Phase 3.2: Analyze existing test coverage
│     └─ Phase 3.3: Identify coverage gaps
│     └─ Returns: test-context-package.json           ← COLLAPSED
├─ Phase 4: Conflict Resolution (conditional)
│  IF conflict_risk >= medium:
│  └─ Inline within Phase 2 context-gathering          ← ATTACHED (3 tasks)
│     ├─ Phase 4.1: Detect conflicts with CLI
│     ├─ Phase 4.2: Log and analyze detected conflicts
│     └─ Phase 4.3: Apply resolution strategies
│     └─ Returns: conflict-resolution.json            ← COLLAPSED
│  ELSE:
│  └─ Skip to Phase 5
├─ Phase 5: TDD Task Generation                       ← ATTACHED (3 tasks)
│  └─ phases/02-task-generate-tdd.md
│     ├─ Phase 5.1: Discovery - analyze TDD requirements
│     ├─ Phase 5.2: Planning - design Red-Green-Refactor cycles
│     └─ Phase 5.3: Output - generate IMPL tasks with internal TDD phases
│     └─ Returns: IMPL-*.json, IMPL_PLAN.md           ← COLLAPSED
│        (Each IMPL task contains internal Red-Green-Refactor cycle)
└─ Phase 6: TDD Structure Validation
   └─ Internal validation + summary returned
   └─ Recommend: plan-verify (external)

Key Points:
• ← ATTACHED: Sub-tasks attached to orchestrator TodoWrite
• ← COLLAPSED: Sub-tasks executed and collapsed to phase summary
• TDD-specific: Each generated IMPL task contains complete Red-Green-Refactor cycle

Error Handling

错误处理

  • Parsing failure: Retry once, then report
  • Validation failure: Report missing/invalid data
  • Command failure: Keep phase in_progress, report error
  • TDD validation failure: Report incomplete chains or wrong dependencies
  • Subagent timeout: Retry wait or send_input to prompt completion, then close_agent
  • 解析失败:重试一次,然后报告
  • 验证失败:报告缺失/无效数据
  • 命令失败:保持阶段为in_progress,报告错误
  • TDD验证失败:报告不完整的链或错误的依赖
  • 子代理超时:重试wait或发送send_input提示完成,然后close_agent

TDD Warning Patterns

TDD警告模式

PatternWarning MessageRecommended Action
Task count >10High task count detectedConsider splitting into multiple sessions
Missing test-fix-cycleGreen phase lacks auto-revertAdd
max_iterations: 3
to task config
Red phase missing test pathTest file path not specifiedAdd explicit test file paths
Generic task namesVague names like "Add feature"Use specific behavior descriptions
No refactor criteriaRefactor phase lacks completion criteriaDefine clear refactor scope
模式警告信息推荐操作
任务数量>10检测到任务数量过多考虑拆分为多个会话
缺少测试修复周期绿阶段缺少自动回滚在任务配置中添加
max_iterations: 3
红阶段缺少测试路径未指定测试文件路径添加明确的测试文件路径
通用任务名称如"Add feature"之类的模糊名称使用具体的行为描述
无重构标准重构阶段缺少完成标准定义清晰的重构范围

Non-Blocking Warning Policy

非阻塞警告策略

All warnings are advisory - they do not halt execution:
  1. Warnings logged to
    .process/tdd-warnings.log
  2. Summary displayed in Phase 6 output
  3. User decides whether to address before execution
所有警告均为建议性 - 不会终止执行:
  1. 警告记录到
    .process/tdd-warnings.log
  2. 在第6阶段输出中显示摘要
  3. 用户决定是否在执行前处理

Error Handling Quick Reference

错误处理快速参考

Error TypeDetectionRecovery Action
Parsing failureEmpty/malformed outputRetry once, then report
Missing context-packageFile read errorRe-run context-gather (workflow-plan-execute/phases/02-context-gathering.md)
Invalid task JSONjq parse errorReport malformed file path
Task count exceeds 18Count validation >=19Request re-scope, split into multiple sessions
Missing cli_execution_idAll tasks lack IDRegenerate tasks with phase 0 user config
Test-context missingFile not foundRe-run phases/01-test-context-gather.md
Phase timeoutNo responseRetry phase, check CLI connectivity
CLI tool not availableTool not in cli-tools.jsonFall back to alternative preferred tool
Subagent unresponsivewait timed_outsend_input to prompt, or close_agent and spawn new
错误类型检测方式恢复操作
解析失败输出为空/格式错误重试一次,然后报告
缺少context-package文件读取错误重新运行context-gather(workflow-plan-execute/phases/02-context-gathering.md)
无效任务JSONjq解析错误报告格式错误的文件路径
任务数量超过18数量验证>=19请求重新划定范围,拆分为多个会话
缺少cli_execution_id所有任务均无ID使用第0阶段用户配置重新生成任务
缺少测试上下文文件未找到重新运行phases/01-test-context-gather.md
阶段超时无响应重试阶段,检查CLI连接性
CLI工具不可用工具不在cli-tools.json中回退到替代首选工具
子代理无响应wait超时发送send_input提示,或close_agent并生成新代理

Post-Execution: TDD Verification

执行后:TDD验证

After TDD tasks have been executed (via workflow:execute), run TDD compliance verification:
Read and execute:
phases/03-tdd-verify.md
with
--session [sessionId]
This generates a comprehensive TDD_COMPLIANCE_REPORT.md with quality gate recommendation.
TDD任务执行完成后(通过workflow:execute),运行TDD合规验证:
读取并执行:
phases/03-tdd-verify.md
,参数为
--session [sessionId]
此操作会生成包含质量门建议的全面TDD_COMPLIANCE_REPORT.md。

Related Skills

相关Skills

Prerequisite:
  • None - TDD planning is self-contained (can optionally run brainstorm before)
Called by This Skill (6 phases):
  • workflow-plan-execute/phases/01-session-discovery.md - Phase 1: Create or discover TDD workflow session
  • workflow-plan-execute/phases/02-context-gathering.md - Phase 2: Gather project context and analyze codebase
  • phases/01-test-context-gather.md - Phase 3: Analyze existing test patterns and coverage
  • Inline conflict resolution within Phase 2 - Phase 4: Detect and resolve conflicts (conditional)
  • compact (external skill) - Phase 4.5: Memory optimization (if context approaching limits)
  • phases/02-task-generate-tdd.md - Phase 5: Generate TDD tasks
Follow-up:
  • plan-verify (external) - Recommended: Verify TDD plan quality and structure before execution
  • workflow:status (external) - Review TDD task breakdown
  • workflow:execute (external) - Begin TDD implementation
  • phases/03-tdd-verify.md - Post-execution: Verify TDD compliance and generate quality report
前置要求:
  • 无 - TDD规划是自包含的(可选择在之前运行brainstorm)
本Skill调用的内容(6个阶段):
  • workflow-plan-execute/phases/01-session-discovery.md - 第1阶段:创建或发现TDD工作流会话
  • workflow-plan-execute/phases/02-context-gathering.md - 第2阶段:收集项目上下文并分析代码库
  • phases/01-test-context-gather.md - 第3阶段:分析现有测试模式与覆盖率
  • 第2阶段内联冲突解决 - 第4阶段:检测并解决冲突(条件执行)
  • compact(外部skill) - 第4.5阶段:内存优化(若上下文接近限制)
  • phases/02-task-generate-tdd.md - 第5阶段:生成TDD任务
后续操作:
  • plan-verify(外部) - 推荐:执行前验证TDD计划质量与结构
  • workflow:status(外部) - 查看TDD任务细分
  • workflow:execute(外部) - 开始TDD实施
  • phases/03-tdd-verify.md - 执行后:验证TDD合规性并生成质量报告

Next Steps Decision Table

下一步操作决策表

SituationRecommended ActionPurpose
First time planningRun plan-verify (external)Validate task structure before execution
Warnings in tdd-warnings.logReview log, refine tasksAddress Red Flags before proceeding
High task count warningConsider new sessionSplit into focused sub-sessions
Ready to implementRun workflow:execute (external)Begin TDD Red-Green-Refactor cycles
After implementationRun phases/03-tdd-verify.mdGenerate TDD compliance report
Need to review tasksRun workflow:status (external)Inspect current task breakdown
情况推荐操作目的
首次规划运行plan-verify(外部)执行前验证任务结构
tdd-warnings.log中有警告查看日志,优化任务处理危险信号后再继续
任务数量过多警告考虑新建会话拆分为聚焦的子会话
准备实施运行workflow:execute(外部)开始TDD红-绿-重构周期
实施完成后运行phases/03-tdd-verify.md生成TDD合规报告
需要查看任务运行workflow:status(外部)检查当前任务细分

TDD Workflow State Transitions

TDD工作流状态转换

workflow-tdd-plan (this skill)
[Planning Complete] ──→ plan-verify (external, recommended)
[Verified/Ready] ─────→ workflow:execute (external)
[Implementation] ─────→ phases/03-tdd-verify.md (post-execution)
[Quality Report] ─────→ Done or iterate
workflow-tdd-plan (this skill)
[Planning Complete] ──→ plan-verify (external, recommended)
[Verified/Ready] ─────→ workflow:execute (external)
[Implementation] ─────→ phases/03-tdd-verify.md (post-execution)
[Quality Report] ─────→ Done or iterate