AI & Agents Skills
Prompting, agent workflows, MCP servers and skill authoring.
3,164 AI & Agents agent skills, including Skill Creator, Mcp Builder, and Discernment Nudge.
3,164 skills

Skill Creator
anthropics
Creates, edits, and benchmarks agent skills, with eval runs, review viewer, and description optimization.

Mcp Builder
anthropics
Guides building high-quality MCP servers in TypeScript or Python, from planning and tool design through testing and evaluation.

Discernment Nudge
anthropics
Appends 2-3 specific follow-up questions to substantive answers so users can check facts, reasoning and missing context.

Plugin Settings
anthropics
Documents the .claude/plugin-name.local.md pattern for storing per-project plugin settings as YAML frontmatter plus markdown.

MCP Integration
anthropics
Guides integrating Model Context Protocol (MCP) servers into Claude Code plugins via .mcp.json or plugin.json.

Hook Development
anthropics
Guides creating Claude Code plugin hooks: event types, prompt-based and command hooks, config formats, output schemas and validation scripts.

Agent Development
anthropics
Guides creating Claude Code plugin agents: frontmatter fields, triggering descriptions, system prompts, and validation.

Claude Opus 4 5 Migration
anthropics
Guides migrating prompts and code from Claude Sonnet 4.0/4.5 or Opus 4.1 to Opus 4.5, updating model strings and optional prompt fixes.

Skill Development
anthropics
Guides creation of Claude Code plugin skills, covering structure, progressive disclosure, writing style and validation.

Plugin Structure
anthropics
Guides the structure, manifest, and component layout of Claude Code plugins.

Command Development
anthropics
Guides creating Claude Code slash commands: Markdown structure, YAML frontmatter, arguments, file references and bash execution.

Skill Creator
anthropics
Creates, edits and benchmarks agent skills, running with-skill vs baseline evals and optimizing skill descriptions.

Build Mcpb
anthropics
Guides packaging a local MCP server with its runtime into a single .mcpb bundle for one-file installation.

Build Mcp Server
anthropics
Guides developers through designing and scaffolding a Model Context Protocol (MCP) server, from use-case discovery to deployment choice.

Build Mcp App
anthropics
Guides building MCP apps: MCP servers that also serve interactive UI widgets rendered inline in chat hosts.

Writing Hookify Rules
anthropics
Guides writing hookify rules: markdown files with YAML frontmatter that match patterns and show messages to Claude.

Plugin Settings
anthropics
Documents the .claude/plugin-name.local.md pattern for storing per-project plugin settings and state with YAML frontmatter.

Mcp Integration
anthropics
Guides integrating Model Context Protocol (MCP) servers into Claude Code plugins via .mcp.json or plugin.json configuration.

Hook Development
anthropics
Guides creating Claude Code plugin hooks: event types, prompt-based and command hooks, config formats, validation and testing.

Agent Development
anthropics
Guides creating Claude Code plugin agents: frontmatter fields, triggering descriptions, system prompts, and validation.

Routines
anthropics
Guides a champion through choosing recurring Claude Code routines and produces ready-to-paste /schedule lines for them.

Patterns
anthropics
Surfaces recurring review/fix patterns and reusable good practices from a repo's recent history, then drafts a rule, skill, or shareable note.

Maturity
anthropics
Scores an organization's AI adoption maturity across seven dimensions and writes a self-contained HTML report with cited evidence.

Climb
anthropics
Guides a champion through assessing where a teammate, themselves, or a team sits on a five-step AI adoption ladder and picking one next move.