The definitive directory for the AI agent ecosystem
Search across 50+ directories. Security-scored. Cross-referenced.
morluto
Reverse engineer anything with agents: one CLI and MCP server to inspect any app, understand it fully down to the binary level.
...moreGitHub Actions
Reverse engineer anything from your terminal or agent with one CLI and MCP server.
GitHub Actions
Reusable model provider adapters and streaming runtime from OpenClaw
Child-scoped report tool over ctx.subagents continuations
Agent-plane presentation selector: composes one agent's tools as Code Mode, native, or both
MCP client bridge: connects to MCP servers and registers their tools on ctx.tools
6a01de90-15be-4d9a-9148-debef44f9576
No description available
cornelius_n8n_io
Blob storage layer for n8n, with filesystem, S3, and Azure Blob backends
rileyhilliard
MCP tools for choosing stacks, creating projects and running their verification checks.
readanydoc contributors
Document parsing and extraction with vision models: an open engine around Claude (or any model), with a Reducto-compatible API
...moretimootten
Small, auditable MCP server for one JMAP mail account (RFC 8620/8621)
sri-chandra-sekhar-a
Temporary package placeholder for staged publishing
ACME's connectors for every capability, shaped for a company MCP gateway, on dummy data.
Skillful.sh aggregates data from every major directory and registry in the AI agent ecosystem into one searchable directory.
We crawl and sync data from 50+ directories, registries, and package managers across the AI agent ecosystem.
Every tool is security-scored, categorized, and cross-referenced across all sources for comprehensive coverage.
Browse, search, and compare tools with detailed profiles showing directory presence, stats, and security grades.
The building blocks powering the next generation of AI applications
Model Context Protocol (MCP) servers are standardized interfaces that let AI models interact with external tools and data sources -- bridging large language models with file systems, databases, APIs, and web services. Originally developed by Anthropic, MCP has become the dominant protocol for extending AI agent capabilities.
AI skills are modular capabilities that extend AI agents and language models. Unlike MCP servers which follow a specific protocol, skills encompass custom tool definitions, function calling schemas, and plugin architectures across multiple platforms -- giving developers a unified view rather than searching dozens of disconnected registries.
Every indexed tool undergoes automated security analysis covering code quality, dependency vulnerabilities, maintenance frequency, license compliance, and known CVEs. AI agents that execute code or install packages on behalf of users introduce supply-chain risk -- security grades help developers quickly identify trustworthy tools for production use.
A single MCP server might be listed on the official registry, published on npm, hosted on GitHub, and curated in multiple community directories. Skillful.sh aggregates and cross-references tools from 55+ sources into one unified directory with live GitHub stats, package registry downloads, and cross-listing data so you can gauge true ecosystem adoption.
From a few hundred MCP servers at the start of 2025, the ecosystem now includes over 581,579 tools spanning MCP servers, AI skills, and autonomous agents. Skillful.sh tracks this growth in real time with ecosystem statistics, trending tools, and growth charts that show how the landscape is evolving week by week.
Browse thousands of MCP servers, skills, and agents. All in one place.