📁文件系统 MCP 推荐

CSS accessibility audit and fixes. Checks focus-visible styles, reduced motion support, contrast preferences, forced-colors mode, visually-hidden patterns, touch target sizes, color contrast ratios, color-only indicators, skip links, and reading order. Use when auditing accessibility, fixing a11y issues, adding focus styles, supporting reduced motion, improving contrast, or making CSS accessible.

Build theming systems with modern CSS. Covers custom properties architecture (primitive, semantic, component tokens), oklch/oklab color spaces, color-mix() for tints and shades, light-dark() for automatic dark mode, @property for typed custom properties, multiple theme support, contrast themes, and forced-colors mode. Use when building a theme, adding dark mode, creating color systems, setting up design tokens, or implementing color schemes.

Modern responsive design using container queries, fluid typography, responsive images, logical properties, viewport units, intrinsic sizing, and preference queries. Prioritizes container queries over media queries. Use when the user asks for responsive CSS, responsive design, mobile layout, fluid typography, responsive images, dark mode, prefers-reduced-motion, or adaptive design.

Upgrade legacy CSS to modern standards. Scans stylesheets for outdated patterns (floats, clearfix, vendor prefixes, old flexbox syntax, px media queries, @import, !important abuse) and replaces them with modern equivalents. Introduces cascade layers, extracts design tokens, and consolidates redundancy. Use when refactoring CSS, modernizing stylesheets, upgrading legacy code, removing vendor prefixes, or replacing floats with grid/flexbox.

Hunt for exploitable, bounty-worthy security issues in repositories. Focuses on remotely reachable vulnerabilities that qualify for real reports instead of noisy local-only findings.

Create CodeTour `.tour` files — persona-targeted, step-by-step walkthroughs with real file and line anchors. Use for onboarding tours, architecture walkthroughs, PR tours, RCA tours, and structured "explain how this works" requests.

Use when: User wants to extend Docker with custom tools, personalize the Docker environment, or set up user-specific Docker customization. Triggers: 'extend docker', 'docker-extend', 'add tools to docker', 'customize docker', 'add my tools to the container', 'personalize docker setup', 'docker user setup', 'install tools in docker'. Does: Interactively sets up Dockerfile.user and docker-compose.override.yml so users can add personal tools to their Docker environment without affecting maintainer files or committing user-specific config to git.

Use when writing or editing any Spacebot copy — README sections, docs, release notes, marketing text, design doc summaries. Covers voice, tone, patterns to avoid, and what good Spacebot writing sounds like.

Use this skill when the user requests to generate, create, or improve documentation for code, APIs, libraries, repositories, or software projects. Supports README generation, API reference documentation, inline code comments, architecture documentation, changelog generation, and developer guides. Trigger on requests like "document this code", "create a README", "generate API docs", "write developer guide", or when analyzing codebases for documentation purposes.

Use when writing or editing documentation in packages/docs. Covers Gitbook markdown syntax, special blocks, page structure, and the SUMMARY.md table of contents. Trigger phrases include "write docs", "add documentation", "docs page", "gitbook", "user manual".

Review CSS for coding style, PatternFly token usage, and best practices. Use when the user asks to review CSS, check styles, or audit CSS files.

**MANDATORY for ALL MCP server work** - mcp-use framework best practices and patterns. **READ THIS FIRST** before any MCP server work, including: - Creating new MCP servers - Modifying existing MCP servers (adding/updating tools, resources, prompts, widgets) - Debugging MCP server issues or errors - Reviewing MCP server code for quality, security, or performance - Answering questions about MCP development or mcp-use patterns - Making ANY changes to server.tool(), server.resource(), server.prompt(), or widgets This skill contains critical architecture decisions, security patterns, and common pitfalls. Always consult the relevant reference files BEFORE implementing MCP features.

Triage, repair, and close MCP attribution issues from the local report API with evidence-driven decisions and isolated Worktrunk worktrees. Use this skill when Codex needs to process `tool` attribution issues and skills-related attribution issues, inspect related runs, decide whether the issue is actionable in `mcp/src` or `config/source/skills`, update attribution fields as `owner=codex`, and then complete the fix loop through GitHub issue tracking, worktree-based code changes, PR submission, and follow-up iteration when the problem is repairable.

This skill should be used when the user asks to "analyze a Dockerfile", "optimize Docker layers", "validate docker-compose", "check container best practices", or "audit Docker configurations".

Mobile app debugging for iOS, Android, cross-platform frameworks. Use for crashes, memory leaks, performance issues, network problems, or encountering Xcode instruments, Android Profiler, React Native debugger, native bridge errors.

Offline-first mobile apps with local storage, sync queues, conflict resolution. Use for offline functionality, data sync, connectivity handling, or encountering sync conflicts, queue management, storage limits, network transition errors.

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

Refactor a large React component file into a directory of smaller files. Use when the user asks to break up, split, or refactor a big component file.

Simplifies code for clarity. Use when refactoring code for clarity without changing behavior. Use when code works but is harder to read, maintain, or extend than it should be. Use when reviewing code that has accumulated unnecessary complexity.

Use when working with Docker containers — debugging container failures, writing Dockerfiles, docker-compose for integration tests, image optimization, or deploying containerized applications

Write unit tests, component tests, and integration tests for AiderDesk using Vitest and React Testing Library. Use when creating new tests, adding test coverage, configuring mocks, setting up test files, or debugging failing tests.

Adds and updates TypeDoc (TSDoc) comments to TypeScript source files in the Medusa codebase. Covers HTTP types, API routes, UI components, data models, service interfaces, JS SDK methods, abstract providers, workflow SDK functions, core-flows workflows and steps, and events. Use when adding TSDoc comments to files in packages/core/types/src/http, packages/medusa/src/api, packages/design-system/ui/src/components, packages/modules/*/src/models, packages/core/types/src, packages/core/js-sdk/src, packages/core/utils/src, packages/core/workflows-sdk/src/utils/composer, packages/core/core-flows/src, or packages/core/utils/src/core-flows/events.ts.

Timeplus streaming SQL covering stream types, EMIT policies, window functions, JOINs, materialized views, external streams, and UDFs. Make sure to use this skill for any SQL-related question including writing queries, debugging SQL errors, understanding streaming behavior, or designing stream processing pipelines, even if the user doesn't explicitly mention streaming SQL.

Design new Claude skills from structured idea specifications. Use when the skill auto-generation pipeline needs to produce a Claude CLI prompt that creates a complete skill directory (SKILL.md, references, scripts, tests) following repository conventions.

Testing strategy for Supabase Studio. Use when writing tests, deciding what type of test to write, extracting logic from components into testable utility functions, or reviewing test coverage. Covers unit tests, component tests, and E2E test selection criteria.

Help write, debug, and explore SQLite database queries using Kysely. Use when writing repository code, exploring data, debugging query issues, or understanding the database schema.

Reviews planning documents for internal consistency -- contradictions between sections, terminology drift, structural issues, and ambiguity where readers would diverge. Spawned by the document-review skill.

Python development principles and decision-making. Framework selection, async patterns, type hints, project structure. Teaches thinking, not copying.

Run and test Python code in a dedicated playground directory. Use when you need to execute Python scripts, test code snippets, investigate CPython behavior, or experiment with Python without affecting the main codebase.

Browser automation CLI using DOMShell MCP server. Maps Chrome's Accessibility Tree to a virtual filesystem for agent-native navigation.

Refactoring assessment and patterns. Use after mutation testing validates test strength (MUTATE phase) to assess improvement opportunities.

Complete guide for reading and editing PDF documents with PyMuPDF.

Manage Docker containers, images, volumes, networks, and Compose stacks — lifecycle ops, debugging, cleanup, and Dockerfile optimization.

Controls a local browser from a sandboxed remote machine. Use when the agent is running in a sandbox (no GUI) and needs to navigate websites, interact with web pages, fill forms, take screenshots, or expose local dev servers via tunnels.

Self-test a mobile feature change or bug fix after implementation in `apps/mobile`. Use this whenever the user asks to verify a mobile change, run simulator acceptance, smoke-test a mobile PR, or provide screenshot proof for a mobile fix. This skill decides between prod vs local API mode, starts the local follow-server when needed, builds a release app, uses Maestro only to bootstrap registration for non-auth work, then switches to screenshot-driven visual validation and returns screenshot evidence.

Provides patterns and code examples for building Next.js 16+ applications with App Router architecture. Use when creating projects with App Router, implementing Server Components and Client Components ("use client"), creating Server Actions for forms, building Route Handlers (route.ts), configuring caching with "use cache" directive (cacheLife, cacheTag), setting up parallel routes (`@slot`) or intercepting routes, migrating to proxy.ts, or working with App Router file conventions (layout.tsx, page.tsx, loading.tsx, error.tsx).

Safely append CSS styles from one file to another, with per-selector grep verification after every append to catch silent truncation that tail or wc-l alone would miss

Debug Python scripts with proper error surfacing and working directory verification

Systematic Python debugging workflow for spreadsheet tasks to isolate environment issues from script logic

Debug Python script execution failures by capturing full output with exit codes and verifying working directory before file operations

Debug Python script execution failures by capturing full tracebacks and verifying working directory

TypeScript compile-checking in projects where npx and node_modules/.bin symlinks fail, with a filtering technique to isolate errors from only newly-created or task-specific files — ignoring pre-existing legacy errors.

Systematic workflow for auditing TypeScript codebases for import hygiene issues (value vs. type imports, duplicates), missing base class extensions, and incomplete barrel file re-exports, with actionable fix patterns.

Create React components with localStorage persistence and keyboard navigation support

Progressive tool-chain PDF extraction with explicit read_file, run_shell, and execute_code_sandbox sequencing

PDF extraction with ordered tool chain: read_file, then run_shell/pdftotext, then execute_code_sandbox/PyMuPDF

Recover PDF text extraction when read_file returns binary data by using pdftotext via shell