coding-framework
Orchestrate the complete coding workflow with multiple development modes (TDD/incremental/spec-dr... Skill: coding-framework Owner: paudyyin Summary: Orchestrate the complete coding workflow with multiple development modes (TDD/incremental/spec-dr... Tags: latest:13.0.0 Version history: v13.0.0 | 2026-08-03T00:55:05.468Z | auto Version 13.0.0 (coding-framework) – Major update with new structure and advanced multi-skill orchestration - Refactored SKILL.md: now includes explicit skill path mapping and extended modular
Rank
62
Safety
84
Downloads
1.0k
Updated
Oct 11, 2026
Version
13.0.0
Source
CLAWHUB
About
What it does, and when to use it.
Capability contract not published. No trust telemetry is available yet. 1K downloads reported by the source. Last updated 10/11/2026.
Avoid when
- Contract metadata is missing or unavailable for deterministic execution.
Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing
Public facts
Every fact links back to the source it came from.
- Vendor
- Clawhubvendor · observed Oct 11, 2026
- Protocol compatibility
- OpenClawcompatibility · observed Oct 11, 2026
- Adoption signal
- 1K downloadsadoption · observed Oct 11, 2026
- Latest release
- 13.0.0release · observed Aug 3, 2026
- Handshake status
- UNKNOWNsecurity
Install and run
Setup complexity: low.
clawhub skill install s172x1s50dh0bgmm3yc9516nr586kk4b:coding-framework- Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
- Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.
Contract: missing
curl -s "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-coding-framework/snapshot"
Run-check
$0.02 USD1 measured facts are behind this paywall: success rate and latency, uptime and estimated cost, when not to use it, how to call it, benchmark scores.
Agents pay $0.02 in USDC. A card payment is $0.50, the smallest a card allows.
Documentation
CLAWHUB
157,620 characters of source documentation, loaded on request.
Extracted files
5 files captured from the source.
SKILL.md
---
name: coding-framework
description: "Orchestrate the complete coding workflow with multiple development modes (TDD/incremental/spec-dr..."
tags: [coding, backend, multi-agent, iterative, template-based]
version: 13.0.0
dependencies:
skills: ["code-review", "brainstorming", "debugging-and-error-recovery"]
---
# Coding Framework �?统一编程框架 v12.2.0
## 你是谁?
你是一个资深编程框架,整合了业界最佳实践:
- Claude Code �?Hook 事件系统和多代理审查
- Claude Plugins Official 的安全审核和渐进式披�?- OpenAI Codex 的标准化代理定义和安全沙�?- Ponytail �?YAGNI 决策阶梯和代码精简哲学
- **Karpathy 四原�?*(v11.5 新增):LLM 编码行为准则
- **Anthropic 官方技能体�?*(v11.6 新增):24 个开发阶段技能路�?---
## Step 0: 开发阶段检测(v11.6 新增�?> 参考:`using-agent-skills` skill 的开发阶段决策树
在开始编码前,先检测当前任务处于哪个开发阶段,加载对应的技能:
```
任务到达
�? ├── 还不知道想要什么? ──────�?brainstorming(设计门�?创意打磨�? ├── 新项�?功能/变更 ────────�?spec-driven-development
├── 有规格,需要任务? ──────�?planning-and-task-breakdown
├── 实现代码 ────────────────�?incremental-implementation
�? ├── UI 工作 ─────────────────�?frontend-design
�? ├── API 工作 ────────────────�?api-and-interface-design
�? ├── 需要更好的上下文? ─────�?context-engineering
�? ├── 需要文档验证的代码�?───�?source-driven-development
�? └── 高风�?不熟悉的代码�?──�?doubt-driven-development
├── 编写/运行测试 ─────────�?test-driven-development
�? └── 基于浏览器? ───────────�?browser-testing-with-devtools
├── 出问题了�?──────────────�?debugging-and-error-recovery
├── 审查代码�?──────────────�?code-review
�? ├── 太复杂? ─────────────�?code-simplifier
�? ├── 安全问题�?───────�?security-and-hardening
�? └── 性能问题�?───�?performance-optimization
├── 提交/分支 ─────────�?git-workflow-and-versioning
├── CI/CD 管道工作 ───────────�?ci-cd-and-automation
├── 弃用/迁移 ─────────�?deprecation-and-migration
├── 写文�?ADR�?───────────�?documentation-and-adrs
├── 添加日志/指标/告警 ───�?observability-and-instrumentation
└── 部署/发布 ─────────�?shipping-and-launch
```
### 阶段检测规�?**检测关键词**�?| 阶段 | 关键�?| 加载技�?|
|------|--------|---------|
| 需求澄�?想法精炼 | 想要什么、想法、概念、变体、头脑风�?| brainstorming |
| 规格定义 | 规格、PRD、需求文�?| spec-driven-development |
| 任务分解 | 任务、拆解、计�?| planning-and-task-breakdown |
| 代码实现 | 实现、编码、开�?| incremental-implementation |
| UI 开�?| UI、界面、组件、样�?| frontend-design |
| API 开�?| API、接口、端�?| api-and-interface-design |
| 测试编写 | 测试、TDD、单元测�?| test-driven-development |
| 调试修复 | 调试、bug、修�?| debugging-and-error-recovery |
| 代码审查 | 审查、review、检�?| code-review |
| 代码简�?| 简化、重构、优化结�?| code-simplifier |
| 安全加固 | 安全、漏洞、加�?| security-and-hardening |
| 性能优化 | 性能、优化、加�?| performance-optimization |
| Git 操作 | 提交、分支、合�?| git-workflow-and-versioning |
| CI/CD | CI、CD、流水线 | ci-cd-and-automation |
| 弃用迁移 | 弃用、迁移、升�?| deprecation-and-migration |
| 文档编写 | 文档、ADR、说�?| documentation-and-adrs |
| 可观测�?| 日志、指标、监�?| observability-and-instrumentation |
| 部署发布 | 部署、发布、上�?| shipping-and-launch |
### 🔴 防跳过机制:Red Flags 表格(v11.7 新增�?> 参考:Superpowers using-superpowers skill �?Red Flags 设计
>
> **核心原则**:即使只�?1% 的可能性适用某plans/README.md
��# Plans Directory Plan files generated by Plan Mode (!j_6) are stored here. Format: plan-YYYYMMDD-HHMMSS.md
_meta.json
{
"ownerId": "kn73z7zhz92tv6skpnmpy5pdfh86j68x",
"slug": "coding-framework",
"version": "13.0.0",
"publishedAt": 1785718505468
}references/agent-system.md
# 代理系统完整文档
> coding-framework 的标准化编程代理定义和调度系统。
## 概述
代理系统提供 7 个专业代理,每个代理有明确的职责、工具权限和输出格式。
灵感来源:Claude Code 代理定义 + Codex YAML 格式 + OpenAI Codex 安全沙箱。
## 代理定义规范
每个代理使用 YAML 文件定义,存放于 `agents/` 目录:
```yaml
name: agent-id # 唯一标识符
display_name: "显示名称" # 中文友好名称
description: "职责描述" # 一句话说明
model: sonnet # 推荐模型
color: green # UI 显示颜色
tools: # 允许使用的工具
- Read
- Grep
trigger_examples: # 触发示例
- "示例消息"
system_prompt: | # 系统提示
详细的系统提示文本...
```
## 代理列表
### code-reviewer(代码审查员)
- **职责**:代码质量、可维护性、最佳实践 + Ponytail 决策阶梯
- **工具**:Read, Grep, Glob, exec
- **输出**:审查报告 + 评分(代码质量/可维护性/最佳实践/精简度)
- **特色**:集成 YAGNI 决策阶梯,识别过度抽象
### security-auditor(安全审计员)
- **职责**:安全漏洞、敏感信息、权限风险 + 25 种安全模式
- **工具**:Read, Grep, Glob, exec
- **输出**:安全报告 + CWE 编号 + 安全评分
- **特色**:覆盖 25 种安全模式,从注入到依赖漏洞
### test-engineer(测试工程师)
- **职责**:测试用例生成、覆盖率分析、测试策略
- **工具**:Read, Grep, Glob, exec
- **输出**:测试用例列表 + 测试代码 + 覆盖率建议
- **特色**:单元/集成/边界/异常全覆盖
### architecture-critic(架构批评家)
- **职责**:模块划分、依赖关系、设计模式、扩展性
- **工具**:Read, Grep, Glob
- **输出**:JSON 结构化发现 + 评分
- **特色**:SRP/开闭原则/循环依赖检测
### performance-analyst(性能分析师)
- **职责**:算法复杂度、资源泄漏、I/O 性能、并发
- **工具**:Read, Grep, Glob, exec
- **输出**:JSON 结构化发现 + 复杂度标注
- **特色**:N+1 检测、内存泄漏识别、可量化优化建议
### maintainability-reviewer(可维护性审查员)
- **职责**:命名质量、函数设计、复杂度、技术债务
- **工具**:Read, Grep, Glob
- **输出**:JSON 结构化发现 + 评分
- **特色**:TODO/FIXME 检测、圈复杂度分析
### documentation-checker(文档检查员)
- **职责**:API 文档、代码注释、README 完整性
- **工具**:Read, Grep, Glob
- **输出**:JSON 结构化发现 + 评分
- **特色**:文档一致性检查、示例代码建议
## 调度逻辑
```
用户请求 → 识别任务类型
→ 关键词匹配(trigger_examples)
→ 文件类型匹配(扩展名)
→ 任务类型匹配(审查/安全/测试)
→ 默认回退:code-reviewer
```
调度优先级:security-auditor > test-engineer > code-reviewer
## 与 sessions_spawn 集成
```
用户请求 → coding-framework 识别任务类型
→ 加载对应 agents/*.yaml
→ sessions_spawn(
model: agent.model,
system_prompt: agent.system_prompt,
task: 用户请求 + 上下文
)
→ 等待子代理返回
→ 格式化输出
```
## 输出格式规范
所有代理输出遵循统一格式:
```markdown
## [代理名称] 审查结果
### 概要
一句话总结发现。
### 详细发现
1. **[严重程度]** 问题描述
- 位置:文件:行号
- 建议:修复方案
### 评分
- 整体评分:X/10
- 置信度:X%
```
## 自定义代理
在 `agents/` 下创建新的 `.yaml` 文件,确保:
- `name` 唯一
- `system_prompt` 足够详细
- `trigger_examples` 覆盖常见场景
- `tools` 最小权限原则references/external-agents.md
# 外部编码代理委派模式
> 来源:coding-agent skill,适配 coding-framework
> 版本:10.1.0(v10.1 新增安全警告和冲突处理)
## 概述
当任务需要外部编码代理(Codex / Claude Code / Pi)时,使用本指南。
**核心原则**:coding-framework 负责质量治理,外部代理负责执行。
---
## 1. 代理选择
| 代理 | 调用方式 | 适用场景 |
|------|----------|----------|
| Claude Code | `claude --print 'task'` | 复杂重构、代码生成 |
| Codex | `bash pty:true command:"codex exec --full-auto 'task'"` | 快速修复、批量任务 |
| Pi | `bash pty:true command:"pi 'task'"` | 轻量任务 |
---
## 2. 调用模式
### Claude Code(非交互)
```bash
# 前台执行(推荐:使用权限白名单)
bash workdir:~/project command:"claude --allowedTools 'Read,Write,Edit' --print 'Your task'"
# 后台执行
bash workdir:~/project background:true command:"claude --allowedTools 'Read,Write,Edit' --print 'Your task'"
```
> ⚠️ **安全警告(v10.1 新增)**:
>
> `--permission-mode bypassPermissions` 绕过所有权限检查,**仅适用于完全隔离的沙箱环境**(如 Docker 容器、临时 VM)。
>
> **不推荐在本地开发环境使用**。生产环境请使用 `--allowedTools` 白名单模式。
### Codex(PTY 模式)
```bash
# 快速一次性任务
bash pty:true workdir:~/project command:"codex exec --full-auto 'Add error handling'"
# 后台长任务
bash pty:true workdir:~/project background:true command:"codex --yolo 'Refactor auth module'"
```
**注意**:Codex 需要 Git 仓库环境,临时任务用 `mktemp -d && git init`。
---
## 3. Git Worktree 并行修复
用于同时修复多个 Issue,每个 worktree 独立工作目录:
```bash
# 1. 创建 worktree
git worktree add -b fix/issue-78 /tmp/issue-78 main
git worktree add -b fix/issue-99 /tmp/issue-99 main
# 2. 在每个 worktree 启动代理
bash pty:true workdir:/tmp/issue-78 background:true command:"codex --yolo 'Fix issue #78: <description>. Commit and push.'"
bash pty:true workdir:/tmp/issue-99 background:true command:"codex --yolo 'Fix issue #99. Commit and push.'"
# 3. 监控进度
process action:list
process action:log sessionId:XXX
# 4. 创建 PR
cd /tmp/issue-78 && git push -u origin fix/issue-78
gh pr create --head fix/issue-78 --title "fix: ..." --body "..."
# 5. 清理
git worktree remove /tmp/issue-78
```
### Worktree 冲突处理(v10.1 新增)
**约束**:worktree 任务必须修改不相交的文件集。
**冲突检测脚本**:
```bash
#!/bin/bash
# check-worktree-conflict.sh
# 检查多个 worktree 是否修改了相同文件
WORKTREES=("$@")
CONFLICT_FILES=""
for wt in "${WORKTREES[@]}"; do
files=$(cd "$wt" && git diff --name-only HEAD 2>/dev/null)
for f in $files; do
if echo "$CONFLICT_FILES" | grep -q "$f"; then
echo "⚠️ 冲突文件: $f (在多个 worktree 中被修改)"
else
CONFLICT_FILES="$CONFLICT_FILES $f"
fi
done
done
if [ -z "$CONFLICT_FILES" ]; then
echo "✅ 无文件冲突"
else
echo "📋 已修改文件: $CONFLICT_FILES"
fi
```
**使用方式**:
```bash
# 启动 worktree 前检查
bash check-worktree-conflict.sh /tmp/issue-78 /tmp/issue-99
# 如果检测到冲突,调整任务分配
```
---
## 4. 完成自动通知
在任务 prompt 末尾追加通知命令:
```bash
codex --yolo exec 'Build a REST API.
When completely finished, run:
openclaw system event --text "Done: Built REST API with CRUD endpoints" --mode now'
```
效果:任务完成后立即触发 wake event,无需轮询。
> **依赖说明**:`openclaw` 命令需要在 PAionUi
Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!
activepieces
AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents
cherry-studio
AI productivity studio with smart chat, autonomous agents, and 300+ assistants.
CopilotKit
The Frontend for Agents & Generative UI. React + Angular
Machine-readable data
The same record, as JSON, for agents and crawlers.
{
"facts": [
{
"factKey": "vendor",
"category": "vendor",
"label": "Vendor",
"value": "Clawhub",
"href": "https://clawhub.ai/paudyyin/skills/coding-framework",
"sourceUrl": "https://clawhub.ai/paudyyin/skills/coding-framework",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-11T18:51:56.978Z",
"isPublic": true
},
{
"factKey": "protocols",
"category": "compatibility",
"label": "Protocol compatibility",
"value": "OpenClaw",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-coding-framework/contract",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-coding-framework/contract",
"sourceType": "contract",
"confidence": "medium",
"observedAt": "2026-10-11T18:51:56.978Z",
"isPublic": true
},
{
"factKey": "traction",
"category": "adoption",
"label": "Adoption signal",
"value": "1K downloads",
"href": "https://clawhub.ai/paudyyin/coding-framework",
"sourceUrl": "https://clawhub.ai/paudyyin/coding-framework",
"sourceType": "profile",
"confidence": "medium",
"observedAt": "2026-10-11T18:51:56.978Z",
"isPublic": true
},
{
"factKey": "latest_release",
"category": "release",
"label": "Latest release",
"value": "13.0.0",
"href": "https://clawhub.ai/paudyyin/coding-framework",
"sourceUrl": "https://clawhub.ai/paudyyin/coding-framework",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-08-03T00:55:05.468Z",
"isPublic": true
},
{
"factKey": "handshake_status",
"category": "security",
"label": "Handshake status",
"value": "UNKNOWN",
"href": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-coding-framework/trust",
"sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-paudyyin-coding-framework/trust",
"sourceType": "trust",
"confidence": "medium",
"observedAt": null,
"isPublic": true
}
],
"events": [
{
"eventType": "release",
"title": "Release 13.0.0",
"description": "Version 13.0.0 (coding-framework) – Major update with new structure and advanced multi-skill orchestration - Refactored SKILL.md: now includes explicit skill path mapping and extended modular loading rules for each sub-skill. - Fully revised opening workflow: distinct separation of brainstorming (design/idea), spec definition, planning, implementation, and review with clear stage detection. - Enhanced Red Flags and design gate processes with step-by-step auto triggers and user checkpoints to prevent process skipping. - Skill dependencies now updated to explicitly include debugging and error recovery by default. - Numerous agent YAMLs, rules, hooks, and references updated for consistency with new orchestration logic. - Removed obsolete skill-card.md and legacy references; all subsystems now documented in the main SKILL.md.",
"href": "https://clawhub.ai/paudyyin/coding-framework",
"sourceUrl": "https://clawhub.ai/paudyyin/coding-framework",
"sourceType": "release",
"confidence": "medium",
"observedAt": "2026-08-03T00:55:05.468Z",
"isPublic": true
}
]
}Record generated Oct 11, 2026.
For crawlers
This page is free to read. The run-check above is the only paid part, and it answers HTTP 402 until it is paid. Everything else here is public.
- One record, as JSON: card, facts, snapshot, contract, trust.
- Every agent, one feed: /.well-known/ai-catalog.json
- What this site sells, and the price: /.well-known/x402
- Paid run-check: /api/v1/agents/clawhub-paudyyin-coding-framework/run-check
