发现 内容与媒体 相关 Skills，比较来源仓库和用途，查看适合你工作流的技能说明。
Perform systematic self-review of code changes before commits using structured checklist. Validates architecture boundaries, code quality, test coverage, documentation, and project-specific anti-patterns. Use b...
Validate and optimize CLAUDE.md files using Anthropic's best practices for focused sessions. Detects contradictions, redundancy, excessive length (200+ lines), emphasis overuse (2%+ density), broken links, and...
Use this skill when writing, editing, or reviewing Python code. Ensures code follows PEP 8 standards and Python best practices.
Review code for correctness, design quality, and complexity using principles from "A Philosophy of Software Design." Use this skill when reviewing pull requests, code changes, diffs, or when asked to find bugs,...
Validates that all imports follow fail-fast principle by detecting try/except ImportError patterns, conditional imports, and optional dependencies. Use before commits, when modifying imports, during code review...
Safe-by-default git workflow and Conventional Commits. Triggers on git operations, reviewing diffs, staging files, creating commits, or branch changes. Prevents destructive operations without consent.
Senior developer code review skill. Use when reviewing code for quality, security, performance, and best practices. Triggers on code changes, PR reviews, or explicit review requests.
執行全面性的程式碼審查，提供具建設性、可落地執行的回饋，涵蓋安全性、效能、程式碼品質、架構設計與測試等面向。
Review implementation quality and regression risks via claude-delegator (Code Reviewer expert). Use after implementation for complex tasks, refactors, or API changes.
Run dual type checking with pyright (fast, development) and mypy (thorough, CI/CD). Use when checking types, before commits, debugging type errors, or validating type annotations. Explains why both tools requir...
CLI review and testing checklist
Review uncommitted code changes against project standards. Similar to /review-code but scoped to changed files only. Suggests fixes and in auto-edit mode fixes automatically.
Logic validation by verbalization. Translate code to plain English to catch semantic errors. Use before running complex logic or when debugging "it should work" failures.
Identifies and fixes ESLint errors and TypeScript type issues across the codebase.
Compares current quality metrics (tests, coverage, type errors, linting, dead code) to baseline and detects regressions. PROACTIVELY invoked after task completion, before marking task complete, or before mergin...
Enforces zero linter errors, common issue prevention, code cleanup. Use when writing code, reviewing code, or ensuring code quality standards.
Captures quality metrics baseline (tests, coverage, type errors, linting) by running quality gates and storing results in memory. PROACTIVELY invoked at feature start, before refactor work, or after major chang...
Run preview build smoke tests and verify runtime config. Use when the user mentions preview smoke, runtime config validation, or pre‑deploy checks.
Automated SRP validation using multi-dimensional detection. Use when reviewing code for SRP compliance, checking "is this doing too much", validating architectural boundaries, before commits, or during refactor...
Detect Single Responsibility Principle (SRP) violations using multi-dimensional analysis. Use when reviewing code for "SRP", "single responsibility", "god class", "doing too much", "too many dependencies", befo...
Reviews code for security vulnerabilities. Use when security concerns are detected or before deploying critical changes.
Run an interactive codebase audit to analyze project structure, detect issues, and generate improvement PRDs. Use when you want to understand a codebase, find technical debt, or plan improvements. Triggers on:...
Enforces Dart doc comments for public APIs, parameter descriptions, usage examples. Use when writing public APIs, reviewing code, or ensuring documentation completeness.
Code review guidelines covering code quality, security, and best practices.
默认按已收录数据中的 GitHub 仓库 Stars 排序。Stars 属于来源仓库，同一仓库的多个技能可能共享相同数值；它反映仓库关注度，不代表单个技能的安装量或质量评分。