Development

Codex MCP 协作流程。触发场景：(1) 新功能开发 (2) 重构 (3) 复杂业务逻辑 (4) 代码审查。提供三阶段协作：需求分析→原型获取→审查，强调批判性思考与只读安全。

Use Codex CLI for sandboxed auditing, debugging, and autonomous prototyping

Ask OpenAI Codex questions about code to understand implementations, architecture, patterns, and debugging. Use when the user asks how code works, where something is implemented, what patterns are used, or needs to understand existing code. Requires Codex CLI installed.

Codex architecture consulting (third brain). Get Codex architecture advice during design phase, forming dual perspective with Claude.

Generates interactive code flow diagrams (.cgraph files). Invoke when users ask: "how does X work", "show me the flow of Y", "visualize the architecture", "diagram the data flow", "create a graph of the codebase", or similar queries about understanding code structure and relationships.

Analyze code for style violations, pattern compliance, and quality issues

Create, manage, and optimize GitHub Codespaces cloud development environments. Configure devcontainers, manage resources, and streamline cloud-based development. Triggers on codespace, devcontainer, cloud IDE, remote development, development environment, container development.

Create, manage, and optimize GitHub Codespaces cloud development environments. Configure devcontainers, manage resources, and streamline cloud-based development. Use when setting up cloud development environments, configuring devcontainers, or managing Codespaces. Triggers on codespace, devcontainer, cloud IDE, remote development, development environment, container development.

Expert code writer - produces clean, production-ready code

AI-powered code review using CodeRabbit CLI and GitHub integration. Use when running local code reviews before commits ("run coderabbit", "review my changes"), configuring .coderabbit.yaml files, integrating CodeRabbit with Claude Code for autonomous review workflows, using @coderabbitai commands in pull requests, or setting up automated code review pipelines.

Integrates CodeRabbit for automated PR code review. Use after PR creation to get AI-powered review feedback before human review.

This skill should be used when CodeRabbit code review feedback needs to be processed and fixed systematically. Use after running `coderabbit --plain` to automatically save feedback, analyze issues using MCP tools, and implement minimal code fixes with proper planning.

Implementation Agent: Full TDD workflow in a single context window. Writes tests, implements code, and iterates until all tests pass.

Self-hosted Convex development in Coder workspaces with authentication, queries, mutations, React integration, and environment configuration

Initial Convex workspace setup in Coder workspaces with self-hosted Convex deployment, authentication configuration, Docker setup, and environment variable generation

CodeQL is a static analysis framework that queries code as a database. Use when you need interprocedural analysis or complex data flow tracking.

Expert-level CodeQL for static analysis, vulnerability detection, and security code scanning

Automate large-scale codebase refactoring using AST transformations on React code.

Comprehensive guide for creating, configuring, and troubleshooting codemagic.yaml files for CI/CD workflows. Use when building mobile apps (Flutter, iOS, Android, React Native) with Codemagic, setting up build pipelines, configuring environment variables, implementing code signing, managing artifacts, or debugging workflow issues.

Analyzes code to identify refactoring opportunities, code duplication, problematic data flows, architectural issues, and code smells. Generates a comprehensive refactoring recommendation plan.

Search and save conversation memories using semantic search. Use this skill when the user mentions past conversations, asks "do you remember", or when historical context from previous sessions would help solve the current problem.

This should only be done if the user asks to "clean up code". At the end of a major change on a git branch this document describes the code clean up tasks that should be performed.

Reviews git history to summarize changes merged while you were away from a consulting project, identifying PRs, code changes, and potential concerns to discuss with the development team.

--- name: codebase-understanding description: Deeply understand entire codebases including file relationships, architecture patterns, and data flows ---