stablyai/orca
remotion-dev/remotion
diegosouzapw/OmniRoute
docker-desktop-windows/docker-desktop
kepano/obsidian-skills MCP · Category
Trending Model Context Protocol repositories
MCP — the Model Context Protocol — went from spec to ecosystem in under a year. Every API, database, IDE, and SaaS now wants an MCP server, and the open-source community is shipping them faster than vendors can keep up. This leaderboard is the wide feed; /best/mcp-servers is the curated cut. The repos here split four ways: single-service connectors (databases, design tools, browsers), registries and meta-servers proxying many backends through one endpoint, horizontal toolboxes bundling file system + shell + search + HTTP for general agents, and protocol-aware runtimes that re-expose themselves as MCP servers so other agents can call them. The category attracts a high ratio of demo-grade work — flashy README, viral hit, no second commit. The score knocks those down by counting sustained mentions, not first-day stars. Reading the leaderboard, recency matters more than usual: the protocol is still evolving and yesterday's reference implementation gets superseded routinely.
Live · top 50 repos · sorted by momentum across 24H
LIVE · 6m| # | Repository | Stars | 24h | 7d | 30d | Trend | Mentions | Actions |
|---|---|---|---|---|---|---|---|---|
| 01 |  diegosouzapw/OmniRoute Never stop coding. Free MIT AI gateway: one endpoint, 290+ providers (90+ free), 500+ models — Kimi, Claude, GPT, OpenAI, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/PWA. Built by 500+ contributors | 31K | +47+0.2% | +10.7K+52.5% | +24.1K+344.8% |     live on X / Twitter | ||
| 02 |  Graphify-Labs/graphify Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store. | 96.4K | +37+0.0% | +4.8K+5.2% | +21.3K+28.3% |  live on Hacker News | ||
| 03 |  koala73/worldmonitor Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface | 74.7K | +29+0.0% | +12.8K+20.6% | +13.8K+22.6% | live on Hacker News | ||
| 04 |  aaif-goose/goose an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM | 51.7K | +55+0.1% | +404+0.8% | +1.4K+2.7% | live on Hacker News | ||
| 05 |  anthropics/claude-plugins-official Official, Anthropic-managed directory of high quality Claude Code Plugins. | 32.7K | +37+0.1% | +318+1.0% | +1.5K+4.9% | live on X / Twitter | ||
| 06 |  headroomlabs-ai/headroom Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server. | 62.6K | +1+0.0% | +1.9K+3.1% | — | live on Hacker News | ||
| 07 |  affaan-m/ECC The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond. | 233.7K | +22+0.0% | +2.4K+1.0% | — | live on Hacker News | ||
| 08 |  farion1231/cc-switch A cross-platform desktop All-in-One assistant for Claude Code, Codex, OpenCode, OpenClaw, Grok Build & Hermes Agent. Only official website: ccswitch.io | 121.4K | -13.5K-10.0% | +2.3K+2.0% | — | live on X / Twitter | ||
| 09 |  xpzouying/xiaohongshu-mcp MCP for xiaohongshu.com | 14.9K | +17+0.1% | +103+0.7% | +453+3.1% | live on X / Twitter | ||
| 10 |  modelcontextprotocol/servers Model Context Protocol Servers | 88.9K | +4+0.0% | +269+0.3% | +1.3K+1.5% | live on Dev.to | ||
| 11 |  tirth8205/code-review-graph Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows. | 26.6K | +1+0.0% | +4.9K+22.6% | +7.7K+40.5% | live on X / Twitter | ||
| 12 |  activepieces/activepieces AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents | 23.4K | +10+0.0% | +89+0.4% | +433+1.9% | live on X / Twitter | ||
| 13 |  PrefectHQ/fastmcp 🚀 The fast, Pythonic way to build MCP servers and clients. | 26.9K | — | +530+2.0% | +1.1K+4.1% | live on Hacker News | ||
| 14 | anthropics/skills Public repository for Agent Skills | 164.3K | +1+0.0% | +1.5K+0.9% | +7.4K+4.7% | live on Hacker News | ||
| 15 |  EverMind-AI/EverOS One portable memory layer for every AI agent: local-first, Markdown-native, user-owned, and self-evolving across apps, tools, and workflows. | 11.6K | +24+0.2% | +185+1.6% | +2.2K+23.2% | live on Hacker News | ||
| 16 |  ChromeDevTools/chrome-devtools-mcp Chrome DevTools for coding agents | 47.6K | +50+0.1% | +382+0.8% | +3.1K+7.0% | live on X / Twitter | ||
| 17 | modelcontextprotocol/inspector Visual testing tool for MCP servers | 10.5K | +7+0.1% | +67+0.6% | +289+2.8% | live on Hacker News | ||
| 18 | anthropics/claude-code Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands. | 139.2K | +2+0.0% | +733+0.5% | +4.6K+3.4% | live on Hacker News | ||
| 19 |  heygen-com/hyperframes Write HTML. Render video. Built for agents. | 37.8K | +1+0.0% | +1.4K+3.9% | — | live on Hacker News | ||
| 20 |  mudler/LocalAI LocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required. | 47.9K | +3+0.0% | +176+0.4% | +715+1.5% | live on X / Twitter | ||
| 21 |  illiahaidar/mcptrustchecker Security scanner for MCP (Model Context Protocol) servers — reads the real published npm/PyPI source, not just metadata, to catch tool poisoning, prompt injection, toxic flows & supply-chain risk. Offline, deterministic A–F Trust Score, SARIF + CI gates. | 179 | +35+24.3% | +103+135.5% | +182 | | ||
| 22 |  mcp-use/mcp-use The fullstack MCP framework to develop MCP Apps for ChatGPT / Claude & MCP Servers for AI Agents. | 10.4K | +27+0.3% | +54+0.5% | +245+2.4% | live on Bluesky | ||
| 23 |  chopratejas/headroom Compress tool outputs, logs, files, and RAG chunks before they reach the LLM. 60-95% fewer tokens, same answers. Library, proxy, MCP server. | 62.6K | -18.4K-22.7% | +1.9K+3.1% | — | live on Hacker News | ||
| 24 |  mksglu/context-mode Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing across 17 platforms via MCP + hooks. | 19.3K | +3+0.0% | +238+1.2% | +1.1K+6.3% | live on Hacker News | ||
| 25 |  github/github-mcp-server GitHub's official MCP Server | 31.7K | +4+0.0% | +171+0.5% | +873+2.8% | live on X / Twitter | ||
| 26 |  czlonkowski/n8n-mcp A MCP for Claude Desktop / Claude Code / Windsurf / Cursor to build n8n workflows for you | 22.4K | +14+0.1% | +60+0.3% | +366+1.7% | live on X / Twitter | ||
| 27 | anthropics/knowledge-work-plugins Open source repository of plugins primarily intended for knowledge workers to use in Claude Cowork | 23.1K | +2+0.0% | +167+0.7% | +1K+4.6% | live on X / Twitter | ||
| 28 |  awslabs/mcp Open source MCP Servers for AWS | 9.5K | +1+0.0% | +31+0.3% | +173+1.9% | live on X / Twitter | ||
| 29 |  ai-boost/awesome-harness-engineering Awesome list for AI agent harness engineering: tools, patterns, evals, memory, MCP, permissions, observability, and orchestration. | 3.3K | +8+0.2% | +98+3.1% | +1K+46.4% | live on X / Twitter | ||
| 30 |  vllm-project/semantic-router Intelligent Mixture-of-Models Router for Efficient Heterogeneous LLMs Inference | 5.1K | +5+0.1% | +61+1.2% | +468+10.2% | live on X / Twitter | ||
| 31 |  NateBJones-Projects/OB1 Open Brain — The infrastructure layer for your thinking. One database, one AI gateway, one chat channel — any AI plugs in. No middleware, no SaaS. | 4.3K | +4+0.1% | +42+1.0% | +338+8.6% | live on X / Twitter | ||
| 32 |  Thysrael/Horizon 📡 Your own AI-powered news radar. Generates daily briefings in English & Chinese. | 用 AI 构建你专属的新闻雷达 | 8.5K | +2+0.0% | +182+2.2% | +1.1K+14.8% | live on Hacker News | ||
| 33 |  yvgude/lean-ctx Control what your AI can see. LeanCTX (Lean Context) is the context intelligence layer for AI agents — one local Rust binary that decides what they read, remembers what they learn, guards what they touch, and proves what they save. 60–90% fewer tokens as the receipt. 76 MCP tools, 30+ agents, local-first. | 3.4K | +11+0.3% | +103+3.1% | +445+15.0% | live on X / Twitter | ||
| 34 |  zhu1090093659/deepseek-pp DeepSeek Web browser extension: AI agent workspace with MCP tools, memory, Skills, automation, web search, and conversation export. | 1.3K | +11+0.8% | +51+3.9% | +349+35.0% | live on X / Twitter | ||
| 35 |  sickn33/agentic-awesome-skills AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 1,987+ agentic skills. Includes CLI, local MCP, catalog, plugins, and Workbench. | 44K | -42-0.1% | +328+0.8% | +2.2K+5.2% | live on X / Twitter | ||
| 36 |  nolabs-ai/nono Sandbox any AI agent in seconds - zero setup, zero latency. | 3.1K | +5+0.2% | +71+2.3% | +304+10.7% | | ||
| 37 |  xuzhougeng/wisp-science Open-source, local-first desktop AI research workbench for scientific computing with Python/R, MCP bioinformatics tools, SSH/WSL/GPU runtimes, and OpenAI/Anthropic models. | 489 | — | +272+125.3% | +489 | | ||
| 38 | anthropics/claude-quickstarts A collection of projects designed to help developers quickly get started with building deployable applications using the Claude API | 17.3K | +7+0.0% | +47+0.3% | +218+1.3% | live on X / Twitter | ||
| 39 |  AnalyseDeCircuit/oxideterm AI-native workspace for local shells and remote machines.Zero Electron, zero OpenSSL, zero telemetry, and no app subscription. | 1.1K | — | +178+18.4% | +290+33.8% | | ||
| 40 |  us/crw Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary. Self-host or use managed cloud. | 488 | +3+0.6% | +49+11.2% | +266+119.8% | live on Hacker News | ||
| 41 |  Ed1s0nZ/CyberStrikeAI The system of action for AI-native cybersecurity—where intent becomes governed execution, evidence becomes operational memory, and every operation improves the next. | 5.3K | +8+0.2% | +84+1.6% | +426+8.7% | live on X / Twitter | ||
| 42 |  kerlenton/mcpsnoop Wireshark for MCP. A transparent proxy that shows every real tool call between your AI client and your MCP servers, live in your terminal. | 297 | +6+2.1% | +21+7.6% | +293+7325.0% | live on Hacker News | ||
| 43 |  builderz-labs/mission-control Self-hosted control plane for AI agents: dispatch tasks, review runs, track spend, and operate OpenClaw, Claude Code, Codex, and other runtimes. | 5.9K | +13+0.2% | +63+1.1% | +5.9K | live on X / Twitter | ||
| 44 |  upstash/context7 Context7 Platform -- Up-to-date code documentation for LLMs and AI code editors | 59.8K | — | +315+0.5% | +1.7K+2.9% | live on X / Twitter | ||
| 45 |  MinishLab/semble Fast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+read | 5.7K | +7+0.1% | +52+0.9% | +301+5.6% | live on Hacker News | ||
| 46 |  stripe/ai One-stop shop for building AI-powered products and businesses with Stripe. | 1.7K | +2+0.1% | +13+0.8% | +76+4.7% | | ||
| 47 |  matlab/matlab-mcp-core-server Run MATLAB® using AI applications with the official MATLAB MCP Server from MathWorks®. This MCP server for MATLAB supports a wide range of coding agents like Claude Code® and Visual Studio® Code. | 1.3K | -6-0.5% | +34+2.8% | +192+18.1% | live on X / Twitter | ||
| 48 |  tavily-ai/tavily-mcp Production ready MCP server with real-time search, extract, map & crawl. | 2.3K | +3+0.1% | +26+1.2% | +115+5.4% | | ||
| 49 |  thClaws/thClaws Open-source AI agent harness in native Rust — GUI, CLI, headless, and webapp from one binary. Multi-provider, MCP, skills, plugins, agent teams. | 1.2K | +1+0.1% | +11+0.9% | +52+4.6% | | ||
| 50 |  ohad6k/emulo Mine your Claude Code and Codex logs into a local you.md agent profile. | 204 | +1+0.5% | +9+4.6% | +203+20300.0% | |
▌ Model Context Protocol — frequently asked questions
What are the best Model Context Protocol projects right now?
As of July 27, 2026, the top-ranked open-source Model Context Protocol projects on TrendingRepo are diegosouzapw/OmniRoute, Graphify-Labs/graphify, koala73/worldmonitor, aaif-goose/goose, and anthropics/claude-plugins-official — ordered by a cross-source momentum score, not raw star count.
What counts as Model Context Protocol?
Model Context Protocol covers protocol servers, connectors, registries, and tooling around MCP ecosystems. TrendingRepo classifies each repository automatically from its GitHub topics, name, description and owner.
How does TrendingRepo rank Model Context Protocol repos?
Each repo gets a 0-100 momentum score combining 24h / 7d / 30d star velocity, fork growth, contributor churn, commit freshness and release cadence, with cross-source mention signals layered on top and anti-spam dampening applied.
How many Model Context Protocol repos does TrendingRepo track?
60 projects are currently tracked in the Model Context Protocol category, refreshed continuously as new repos break out across the signal sources.
How often is this list updated?
Roughly every 20 minutes. Automated collectors re-scan GitHub, Hacker News, X, Bluesky, Product Hunt, Dev.to and more, then recompute the rankings — so the leaderboard reflects momentum within the last hour.