Skill Relationships
Skill → Skill usage relationships, with cross-layer references. Same color blocks indicate the same skill form.
Skill Composition
Each skill is designed for a specific kind of caller — some only fire when you name them, others load automatically when the model judges them relevant. The split below summarises both modes at a glance.
These skills only fire when you call them by name. The agent will not load them on its own — they stay inert until invoked.
These skills are loaded automatically by the agent when it judges the context relevant. You usually don't need to call them yourself.
Skill List
Guides creation of autonomous agents for Claude Code plugins. Provides frontmatter specs, system prompt design, triggering examples, and validation scripts. Outputs well-structured agent files with clear trigger conditions and system prompts, enabling reusable autonomous capabilities in plugins.
One-shot migration skill that updates code, prompts, and API calls from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5. It searches and replaces model strings per platform, removes unsupported beta headers, adds effort param, and applies optional prompt fixes for known behavioral diffs. Value: fast, safe model upgrade with minimal manual edits.
A skill for developing Claude Code slash commands. It teaches command Markdown format, YAML frontmatter fields, dynamic arguments, file references, and bash execution. Output is reusable, well-structured commands that standardize workflows and improve team efficiency.
Guidance for distinctive, intentional frontend visual design when building or reshaping UI. Uses a design-lead approach: ground in subject, define token system (color/type/layout/signature), brainstorm-plan-critique-build loop, avoid templated defaults, deliberate copy. Produces unique, accessible, non-generic UI with a memorable signature element.
Provides comprehensive guidance for creating and implementing Claude Code plugin hooks, covering prompt-based and command hooks, all hook events, configuration formats, matchers, and security best practices. Enables event-driven automation to validate tool use, enforce policies, load context, and integrate external tools via hooks.json and settings.
Guides integration of Model Context Protocol servers into Claude Code plugins to connect external services as tools. Covers .mcp.json/plugin.json config, server types (stdio/SSE/HTTP/ws), auth, tool naming, and security. Delivers verified, portable MCP setups with automatic external tool access.
Documents the `.claude/plugin-name.local.md` pattern for storing per-project plugin config with YAML frontmatter and markdown. Provides bash parsing techniques, hooks/commands/agents reading methods, and best practices. Enables user-configurable, restart-aware plugin behavior without code changes.
A style guide skill for Claude Code plugin development. It documents standardized directory layout, plugin.json manifest config, component organization (commands, agents, skills, hooks, MCP), portable ${CLAUDE_PLUGIN_ROOT} path usage, naming conventions, and auto-discovery. Value: enables developers to scaffold maintainable, seamlessly integrated plugins with correct structure.
This skill guides creation of effective Claude Code plugin skills. It uses example-driven analysis to plan reusable scripts/references/assets, scaffolds the directory manually, and writes SKILL.md with third-person triggers and imperative style. Output is a lean, self-contained skill with progressive disclosure that extends Claude's capabilities reliably.
A skill for writing Hookify rules in Claude Code. It teaches the markdown+YAML rule file format, event types (bash/file/stop/prompt), regex patterns, and condition operators. Users create .claude/hookify.*.local.md files to watch behaviors and show warnings or block actions, improving safety and process compliance.
Installation
Copy and run in your terminal to add anthropics/claude-code to your Agent environment.
npx skills add anthropics/claude-code