Skill Library
AI capabilities from across the web (MCP / Agent Skills), 1208 in total · some can become online tools
fess
Open-source enterprise-grade search engine supporting multiple data sources and languages.
LetsFG
Provides flight and hotel search and booking capabilities for AI agents.
crw
Lightweight web scraping, crawling, and search API with MCP server support.
crawl4ai
An open-source web crawler and scraper tool that converts any website into clean, LLM-friendly Markdown.
hyperresearch
Transform Claude Code or Codex into a deep research agent that collects, searches, and synthesizes online research.
llm-wiki-compiler
Compile raw materials into interconnected wikis for use by both AI and humans.
new-term-radar
Daily tracking of emerging AI terms, including models, products, and repositories.
JobSpy
A library for scraping job listings from multiple job boards.
search2ai
Self-hosted AI agent search gateway for unified search, news, and page extraction.
open-seo-mcp-skills
Open-source SEO and GEO skills, powered by real GSC/GA4/ad data.
OpenOSINT
AI-driven OSINT agent with interactive REPL, MCP server, and CLI support.
zotero-mcp
The Zotero MCP plugin enables deep integration between AI assistants and Zotero via the MCP protocol.
SurfSense
SurfSense is an open-source AI agent network research platform providing real-time data connectivity via REST API or MCP server.
web-search-mcp
Locally hosted Web search MCP server supporting multiple search engines and content extraction.
mcp-searxng
Provides private web search capabilities for AI assistants via SearXNG.
yargi-mcp
MCP server for the Turkish legal database, supporting queries across multiple legal resources.
12306-mcp
A 12306 ticket information search server based on the MCP protocol.
fetcher-mcp
MCP server for scraping web content using Playwright headless browser.
brave-search-mcp-server
MCP server integrating Brave Search API, providing comprehensive search capabilities.
duckduckgo-mcp-server
An MCP server providing web search and content scraping/parsing capabilities via DuckDuckGo.
open-webSearch
Multi-engine MCP server for proxying web search and content retrieval.
tavily-mcp
Production-grade MCP server for real-time search, extraction, mapping, and crawling.
modelcontextprotocol
Official MCP server implementation for the Perplexity API platform, providing real-time web search, reasoning, and research capabilities.
bb-browser
Control Chrome via real browser state, enabling AI agents to directly use logged-in sessions.