Concepts/mcp
MCP - an access layer for AI agents
Model Context Protocol - an open protocol that lets an agent connect to the outside world: documentation, repositories, the browser, databases, and local files. An MCP server is an outlet with tools that you give the agent; the client (Claude Code, Cursor, Codex) decides which tool to call and when.
Servers
14,522
official MCP registry
With GitHub data
10,199
stars, recency, license
Snapshot
Jul 02
updated daily
Score
v0
open signals, no FUD
The MCP market map lives at /mcp: a full catalog, score, and machine-readable layer.
Open market mapThe gist
Why an agent needs MCP
Without MCP, an agent sees only the code in the repository. With MCP, it reads up-to-date documentation (context7), controls the browser (playwright), works with issues and PRs (github), and accesses databases and internal systems. One protocol instead of N integrations: write a server once and it works in any MCP client.
#1 · vendor
mcp-server
Puter MCP enables AI tools to interact with Puter: manage files, websites, workers, and more
#2 · community
Netdata
Real-time infrastructure monitoring with metrics, logs, alerts, and ML-based anomaly detection.
#3 · vendor
Oh My Posh Validator
Validate oh-my-posh configurations and segment snippets against the official schema.
#4 · community
Scrapling MCP Server
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.
#5 · vendor
Agent Skills Search Server
Search and discover Agent Skills from the skills.sh registry. Powered by HAPI MCP server.
#6 · community
claude-flow
AI orchestration with hive-mind swarms, neural networks, and 87 MCP tools for enterprise dev.
#7 · community
Context7
Up-to-date code docs for any prompt
#8 · community
tldraw
Draw and visually collaborate with your agents on tldraw's canvas.
Data source: the official MCP registry (registry.modelcontextprotocol.io) + GitHub API. The score methodology is open, with an honest disclaimer.