Codex First Executor
Fork Overlay：Codex-first 外部执行器策略。按任务特征路由到 Claude 或 Codex，Codex-first（非品牌路由）。使用时机：需要决定是否将当前任务派发给 Codex 时加载此 skill。
Claude Permissions Optimizer
Optimize Claude Code permissions by finding safe Bash commands from session history and auto-applying them to settings.json. Can run from any coding agent but targets Claude Code specifically. Use when experiencing permission fatigue, too many permission prompts, wanting to optimize permissions, or needing to set up allowlists. Triggers on "optimize permissions", "reduce permission prompts", "all…
Review Contract
Fork Overlay：Review Contract 三档 Tier 分类 + Anti-Leniency 注入。在 ce:review 基础上增加 Blocking/Analytical/Advisory 分层和结论类型系统。使用时机：运行 ce:review 前阅读此协议以理解审查标准。
User First Design
This skill should be used when designing user interfaces with a "worst user first" philosophy. It applies when creating forms, pages, flows, or any user-facing interface where usability and emotional comfort are priorities. Generates structured UI design proposals with interaction flows, layouts, copy guidelines, and emotion management strategies.
Ce:compound
5: 沉淀经验，构建团队知识库
Ce:commit
提交当前改动：检查状态、生成清晰 commit message，并创建单个 git commit
Ce:pr
创建 PR 并询问是否合并到主分支
Ce:product Pulse
生成产品脉搏报告：按时间窗口汇总用户体验、使用情况、质量、错误和待调查信号
Intent Gate
Fork Overlay：ce:work 意图分类门控。在执行前识别任务意图（实现/修复/重构/探索），设定对应的执行策略。使用时机：ce:work Phase 0（环境扫描）之后、Phase 1（Quick Start）之前。
Ce:setup
诊断并配置 compound-engineering 环境：检查 CLI 依赖、插件版本和仓库本地配置
Ui Review Contract
Fork Overlay：UI Review Contract — 四信号门触发协议。自动检测任务是否涉及 UI，决定是否在 ce:review 中加入视觉审查层。使用时机：运行 ce:work 或 ce:review 时，系统自动检查四信号门以决定是否触发 design review。
Deploy Docs
Validate the MkDocs documentation site and GitHub Pages deployment for the public repository
Workflows Brainstorm
探索需求和方案，支持 [P]/[P+] 多视角讨论，并写出可继续规划的 brainstorm 文档
Agent Native Audit
Run comprehensive agent-native architecture review with scored principles
Agent Native Architecture
Build applications where agents are first-class citizens. Use this skill when designing autonomous agents, creating MCP tools, implementing self-modifying systems, or building apps where features are outcomes achieved by agents operating in a loop.
Ce:ideas
管理想法停车场 IDEAS.md：无参数时展示并推荐已有想法，有参数时生成新方向
Ce:ideate
0: 生成有根据的改进方向，在 brainstorm 之前使用
Ce:doctor
健康检查：检测 CLI、MCP、认证状态，支持一键修复
Ce:release Notes
查询插件发布说明：总结最近版本变化，或回答某个 skill/版本何时变更
Creating Agent Skills
Expert guidance for creating, writing, and refining Claude Code Skills. Use when working with SKILL.md files, authoring new skills, improving existing skills, or understanding skill structure and best practices.
Ce:work Beta
[Beta] 执行计划并支持外部委派：在 ce:work 基础上实验 Codex 委派以节省上下文
Patch Approval
Fork Overlay：Codex Patch Approval 咨询版。当 Codex 返回 patch 时，Claude 审批后才写入文件。使用时机：Codex 以 patch/diff 格式返回代码变更时，由 Claude 作为审批层。
Ce:resume
回归项目入口：读取最近 git log、IDEAS.md 和 active plan，输出三段摘要
Executor Capability Gate
Fork Overlay：外部模型调用前置检查门控。调用 Codex/Gemini 之前运行五项检查，防止调用失败浪费时间。使用时机：任何调用外部模型（Codex [C]、Gemini [G]）之前自动运行。
Ce:simplify Code
简化近期改动：在保持行为不变的前提下提升代码清晰度、复用性、质量和效率
Ce:commit Push Pr
提交、推送并创建 PR：生成价值导向 PR 描述，也支持仅刷新 PR 描述
Dhh Rails Style
This skill should be used when writing Ruby and Rails code in DHH's distinctive 37signals style. It applies when writing Ruby code, Rails applications, creating models, controllers, or any Ruby file. Triggers on Ruby/Rails code generation, refactoring requests, code review, or when the user mentions DHH, 37signals, Basecamp, HEY, or Campfire style. Embodies REST purity, fat models, thin controlle…
Document Review
Review requirements or plan documents using parallel persona agents that surface role-specific issues. Use when a requirements document or plan document exists and the user wants to improve it.
Workflows Review
对代码、分支或计划进行多角度结构化审查，输出 findings、风险与建议
Changelog
Create engaging changelogs for recent merges to main branch
Sync Targets
A Claude skill from Jerrylalala/compound-engineering.
Ce Work Integration
Fork Overlay：Task Bundle 持久化 + Failure FSM 集成。在 ce:work 基础上增加 state.md 读写和状态机转换。使用时机：执行 ce:work 时，如果任务有对应的 Task Bundle（docs/tasks/<id>/），加载此 skill 以启用持久化和状态追踪。
Ce:plan
2: [T] 制定实施计划，生成执行合约
Task Bundle
管理 Task Bundle 工件结构（docs/tasks/），提供创建、查询、状态更新的规范
Compound Promotion Ladder
Fork Overlay：经验分层沉淀升级阶梯。检测是否值得将 solution 升级为 pattern 或 skill。使用时机：ce:compound 完成后手动调用，分析 docs/solutions/ 中的重复模式。（不会自动触发，需主动加载此 skill）
Workflows Plan
基于需求或 brainstorm 产出 Claude 可执行、风险明确、任务原子化的计划文档
Ce:work
3: [T][T+][V][V+][R] 执行计划（Agent Teams+四层验证）
Ce:sync Upstream
[F] 检测上游仓库更新，生成智能分析报告，辅助决策是否整合
Agent Browser
Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to…
Dspy Ruby
Build type-safe LLM applications with DSPy.rb — Ruby's programmatic prompt framework with signatures, modules, agents, and optimization. Use when implementing predictable AI features, creating LLM signatures and modules, configuring language model providers, building agent systems with tools, optimizing prompts, or testing LLM-powered functionality in Ruby applications.
Ce:polish Beta
[Beta] 启动开发服务器并在浏览器中体验功能，迭代打磨 UI、交互和细节
Andrew Kane Gem Writer
This skill should be used when writing Ruby gems following Andrew Kane's proven patterns and philosophy. It applies when creating new Ruby gems, refactoring existing gems, designing gem APIs, or when clean, minimal, production-ready Ruby library code is needed. Triggers on requests like "create a gem", "write a Ruby library", "design a gem API", or mentions of Andrew Kane's style.
Ce:strategy
创建或维护 STRATEGY.md：沉淀产品问题、方案、用户、指标和工作主线
Ce:sessions
检索历史会话：跨 Claude Code、Codex、Cursor 查询过去做过什么、尝试过什么和如何排查
Ce:update
|
Ce:compound Refresh
刷新经验库：检查 docs/solutions/ 中过期、漂移或重复的经验文档，并更新、合并、替换或标记
Ce:review
4: [C][G][T] 多代理代码审查 + autofix
Ce:brainstorm
1: [P/P+][C][G][R] 探索需求，生成需求文档
Team Mode
Multi-agent collaboration overlay for ce:plan/work/review. ce:brainstorm structured convergence is built-in to [P] mode, not via [T]. Enables contract-gated execution with single writer principle, real Agent Teams verifier, and deterministic patch gate.