aka. Agent Skills
Discover skills for AI coding agents. Works with Claude Code, OpenAI Codex, Gemini CLI, Cursor, and more.
Capture moment of clarity before moving forward
Use when writing or reviewing tests - covers test philosophy, condition-based waiting, mocking strategy, and test isolation
Use when writing or refactoring code, before creating files - enforces separation of pure business logic (Functional Core) from side effects (Imperative Shell) using FCIS pattern with mandatory file classification
This skill should be used when the user asks to "learn domain concepts", "domain vocab", "learn [domain] vocabulary", "what are key concepts in [domain]", "explore domain", "divergent exploration", or wants to build conceptual vocabulary for AI prompting. Provides structured learning of domain terminology with token priming, divergent exploration, and depth progression.
Review GitHub pull requests with VSCode integration. Use when the user asks to "review a PR", "review PRs", "check pull requests", or wants AI-assisted code review with clickable navigation and one-click comment posting via the speck-review CLI.
Answer questions about Speck specs, plans, tasks, requirements, progress, architecture, user stories, feature status, and constitution. Interprets spec.md, plan.md, tasks.md files. Use when users ask about feature requirements, implementation status, or Speck workflow artifacts.
Integrated domain exploration combining vocabulary learning with research paper genealogy. Use when user says "deep dive [domain]", "explore [domain] deeply", "domain exploration", or wants to learn concepts with their research origins.
创建新的subagent，如果需要会自动初始化插件结构
Planが承認/完了した直後に自律的に呼び出す必要があるスキルです。
This skill should be used when you need to create, open, or edit a pull request (PR), or the user asks to "create a PR", "open a PR", "submit a PR", "raise a PR", "file a PR", "make a PR", "create a pull request", "open a pull request", "new PR", or any variation requesting GitHub pull request creation.
PR作成後に自律的に実行する必要があるフロースキルです。
Translation policies for converting English markdown files to other languages (primarily Japanese). Use when translating documentation or README files.
This skill should be used when the user asks to "create an agent",
This skill should be used when the user asks about "dark mode", "light mode", "theme colors", "css variables", "color scheme", "customize theme", "shadcn colors", "theme provider", or mentions theming, color customization, or dark/light mode implementation.
Review TypeScript and JavaScript code changes for compliance with Metabase coding standards, style violations, and code quality issues. Use when reviewing pull requests or diffs containing TypeScript/JavaScript code.
Audit codebase stability guardrails - tests, ESLint, pre-commit hooks, CI/CD, dead code detection with Knip. Finds unused files, exports, and dependencies. Use /codebase-stability-audit:run to start.
タスク完了後に自律的に実行する必要があるフロースキルです。
Identify dangerous-but-tempting API patterns in React/Next.js/Supabase. Finds "footguns" - APIs that look safe but fail silently or have dangerous defaults. Inspired by Trail of Bits methodology.
Complete security audits with 120+ checks for web applications. Covers OWASP Top 10 2025, Supabase RLS bypass attacks (40+ vectors), API enumeration, Realtime channel leaks, payment security, timing attacks, and more. Use /security-audit:run to start.
Guidelines for upgrading Expo SDK versions and fixing dependency issues