wasabeef/claude-code-cookbook

Analyze Dependencies evaluates project dependencies and architectural health, generating dependency matrices, fan-in/fan-out analysis, and prioritized improvement suggestions. Use it to detect circular dependencies, layer violations, excessive coupling, and orphaned modules.

Analyze Performance evaluates Core Web Vitals and quantifies user experience improvements from optimizations. Use it when analyzing page speed, identifying performance bottlenecks, or prioritizing performance fixes.

Check Fact verifies whether statements are accurate by checking a project's code, documentation, and config files. Use it to fact-check technical claims, architecture choices, or documentation accuracy.

Check Prompt evaluates and improves AI prompt quality by analyzing prompts against best practices for structure, executability, prohibited items, quality assurance, roles, and continuous improvement. Use it to score, deep-analyze, or automatically fix prompts for AI agents.

Commit Message generates commit messages from staged changes (git diff --staged) and copies the top suggestion to the clipboard. Use it when you want to quickly write a conventional commits message without running git commands.

Context7 searches technical documentation using the Context7 MCP, triggered by requests like "check the docs" or "how to use this library?". Use it to look up API references and solutions from official docs within Claude Code.

Design Patterns suggests applicable software design patterns and checks code for SOLID compliance and anti-patterns. Use it to evaluate existing architecture, propose refactors, and get example implementations.

Explain Code provides detailed explanations of code behavior, logic, patterns, performance issues, and security concerns. Use it when asked to explain this code or understand how a function, algorithm, or project works.

Fix Error analyzes error messages to identify root causes, predict resolution time, and suggest proven fixes. Use it when you have build errors, test failures, or stack traces and need immediate or preventive solutions.

Issue Priority Ranking lists open issues and sorts them by priority using labels, assignees, and summaries. Use it when you need to see the most important issues in a repository or create a PR from an issue.

Multi-Role Analysis runs multiple specialized review roles (like security, performance, or frontend) in parallel against the same target and merges their findings into an integrated report. Use it when you want a multifaceted evaluation of code, plans, or designs in one command.

PR Auto Update analyzes Git changes to automatically update pull request descriptions and labels. Use it to keep PR metadata accurate after code changes, with options for dry-run and selective updates.

PR Checks monitors GitHub Actions CI status and tracks pull request checks until they complete, so you can wait for all checks to pass before merging.

PR Create analyzes Git changes and creates draft pull requests with generated descriptions, auto-selected labels, and preserved PR templates. Use it to open a PR without manually filling in details.

PR Fix helps developers efficiently respond to pull request review comments by classifying feedback by priority and using a 3-stage error analysis approach to identify root causes. Use it to address PR feedback, implement review fixes, and verify code quality.

PR List displays open pull requests in the current repository and ranks them by priority based on type, draft status, and labels. Use it when you need to see which PRs need attention first.

PR Review systematically reviews pull requests for code quality, security, performance, and architecture using prioritized comment templates and constructive feedback.

Plan Mode creates detailed implementation strategies before coding, including requirements, design, and task breakdown, and waits for explicit approval before starting work.

Refactor performs safe, incremental code refactoring with quantitative SOLID principles evaluation, visualizing technical debt and clarifying improvement priorities.

Role Debate triggers a structured multi-role discussion in Claude Code to explore trade-offs and reach integrated conclusions. Use it when you need to balance competing perspectives like security vs performance or architecture choices.

Role Help guides users in choosing the right Claude Code role for their situation, covering security, architecture, performance, frontend, mobile, review, QA, and multi-role comparisons.

Role Switching lets a Claude Code agent adopt expert personas like security auditor, performance optimizer, code reviewer, architect, or QA engineer to perform specialized, evidence-based analysis. Use it when you need a focused audit, review, or optimization with structured professional output.

Screenshot captures macOS screenshots (window, full screen, or region) and lets Claude analyze the image for UI, error, code, and data visualization issues. Use it when you need to visually inspect or get feedback on anything displayed on screen.

Search Gemini runs web searches through the Gemini CLI to get latest information. Use it when an agent needs current data, error solutions, or technology comparisons instead of Claude's built-in WebSearch.

Semantic Commit analyzes working tree changes and proposes logical commit splits with semantic messages, then commits each group after approval. Use it when a changeset spans multiple features, fixes, or docs.