MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. Each candidate is checked by a language model from its name, description and topics, and only actual MCP servers are listed (not clients, SDKs, lists or skills). Package metadata comes from ecosyste.ms (CC BY-SA 4.0). A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
her-eyesgithub.com/Penty-d/her-eyesupdated this year
带上她的眼睛 · Give a text-only LLM eyes — a single-binary MCP tool that lets agents like Claude Code / Codex call a vision model to extract structured key information from images.
MIT ★ 2 -
A distributed LLM architecture combining local language models, Model Context Protocol (MCP), and Azure Table Storage. Features OpenAI-compatible APIs, tool calling, and real-time streaming for AI-powered data querying.
-
MCP server exposing a RAG service as tools a language model can call — search, read, stats — with tested error messages and cross-repo integration tests.
MIT -
MCPSERVERgithub.com/adinath09/MCPSERVERupdated this year
This project demonstrates an intelligent AI agent built using LangGraph that can dynamically select and call multiple tools via the MCP (Multi-Call Protocol). The agent integrates asynchronous execution, external APIs, and large language models (Groq) to process user queries efficiently.
-
selenium-mcp-servergithub.com/mm541/selenium-mcp-serverupdated this year
It is a standalone Python application that exposes a comprehensive suite of Selenium browser automation commands as a FastMCP server. This allows a language model, AI agent, or any other client capable of calling tools to control a web browser, perform complex web scraping, and automate web-based tasks by issuing natural language commands.
-
Agentic AI system for enterprise incident resolution using LLM agents, RAG, tool calling, MCP, and governed execution.
-
LangGraph agent over the Bitext support dataset: LLM routing, tool calling, SQLite memory, MCP server, Streamlit UI
-
Inventory management AI agent built with Spring AI 2.0, Spring Boot 4 and a local LLM (Ollama). Memory, tool calling, MCP and RAG.
-
spring-ai-mcp-integrationgithub.com/gvj116/spring-ai-mcp-integrationupdated this year
MCP (Model Context Protocol) integration with Spring Boot and Spring AI — Java tools auto-called by LLM via natural language
MIT -
PokéDex con IA: agente con tool calling escrito a mano, servidor MCP, 3 LLMs con fallback (Groq → Gemini → Anthropic), evals y visión. Next.js + Supabase (RLS) + FastAPI.
MIT -
Demo: stop prompt injection on AI agent tool calls with an MCP server that authorizes every action via renker-core-authz. Runnable, no LLM or API key needed.
Apache-2.0 -
A lightweight HTTP/1.1 server written in C — prefork event loop, CGI/FastCGI, React SSR — with a built-in native C LLM agent runtime (tool calls, MCP, skills, memory, ReAct).
MIT ★ 1 -
MCP server exposing backend tools (database lookup, ticket creation, notifications) for LLM tool-calling.
-
mseep-databutton-app-mcpstale
Call your Databutton app endpoints as LLM tools with MCP
pypi MIT -
ollama-agent-kitgithub.com/niceunderground/ollama-agent-kitupdated this year
Minimal Node.js agent loop for Ollama: local LLMs that read/write/edit files, run shell commands, and call MCP tools. MCP client and server, two dependencies, fully offline.
MIT ★ 2 -
MCP Server with agentic tools to migrate a docker containerized project to kubernetes. Run & call it through claude desktop or any LLM client.
-
llm-mcp-fastapigithub.com/ahmadalsharef2/llm-mcp-fastapiupdated this year
FastAPI + MCP server for LLM inference — chat API with tool-calling, streaming responses, and model-context-protocol support.
★ 1 -
expense-trackergithub.com/Krishna-web-hub/expense-trackerupdated this year
FastMCP-based expense tracking MCP server backed by SQLite — add/query/categorize expenses as tools an LLM agent can call.
-
Live Berlin public-transport data platform and the LLM agent that queries it: a hand-written tool-calling loop, an MCP server and client, a LangGraph rebuild, and hybrid pgvector retrieval, over read-only SQL with every tool call audited.
MIT -
cryptospect-cligithub.com/afshinator/cryptospect-cliupdated this year
A portable CLI tool that fetches live cryptocurrency data, computes high-signal market regime metrics, and outputs clean JSON — optimized for AI agents, LLM tool-calling, and MCP workflows.
MIT -
ARES agent server with multi-provider LLM support, tool calling, RAG, and MCP integration
cargo MIT -
Grounded enterprise ontology agent with Neo4j, typed retrieval, LLM tool calling, evaluation, and read-only MCP tools.
-
mup-mcp-servergithub.com/ricky610329/mupupdated this year
MCP server that turns HTML MUP panels into interactive tools for LLMs — render UI, call functions, and pipe data between panels
npm MIT -
x402farm-mcpgithub.com/laurenthalbrun/x402farm-mcpupdated this year
MCP server exposing pay-per-call tools to LLM clients. JSON-RPC over Streamable HTTP, listed in the official Model Context Protocol registry.
-
Local MCP server exposing the Servico CRM tenant API (api/v1) as typed tool calls for LLM agents like Claude Desktop and Cursor.
npm MIT -
💬 Model Context Protocol (MCP) Server for LLM Agentic WhatsApp Automation & Tool Calling
-
Voice-preserving SEO content MCP server — 16 rule-based tools, 7 languages, no external LLM calls
MIT -
findatamcpgithub.com/hanlinlibham/findatamcpupdated this year
Feed LLMs financial data and interactive charts in one call — an MCP server for A-share quotes, financials, and macro indicators, 42 tools with results-as-UI.
MIT ★ 1 -
MCPTap is a lightweight LLM proxy for AI agents that intercepts traffic (currently over 200 providers) and controls MCP tool calls / protects files from AI agent.
NOASSERTION ★ 1 -
Control Apache Airflow with natural language via MCP. Chat with your workflows using Claude, GPT, or any LLM — no REST API calls needed. Supports Airflow 2.x (43 tools) & 3.0+ (45+ tools).
MIT ★ 54 -
Containerized FastAPI backend where an LLM agent performs CRUD through tool calls. Layered architecture, Docker Compose.
-
MCP-Servergithub.com/kiranimmadi2/MCP-Serverupdated this year
MCP-Based AI Code Analysis Tool: Enables LLM agents to interact with, analyze, and debug codebases via structured MCP tool calls — scanning project structure, extracting code elements, and detecting Python anti-patterns.
★ 1 -
MCP server exposing NapCat / OneBot v11 QQ-bot APIs as tools, plus an LLM-driven QQ bot with memory, KB, tool-calling and web search
MIT -
Streaming LLM chat over governed data. The model never sees a row — it calls typed tools over a profiled catalogue, and every call is audited to Kafka before its result is served.
MIT -
AI customer-support agent built on an MCP server and n8n workflows, with LLM tool calling
-
Multi-agent AI disaster relief orchestrator using LLM tool calling, MCP, FastAPI, and real-time operational data.
MIT -
energentgithub.com/ritesh-raushan/energentupdated this year
AI-powered Building Energy Optimization Platform using EnergyPlus, MCP Tool Calling, FastAPI, React, and Open-Source LLMs.
-
ReasonForge-MCP-Servergithub.com/RoyCoding8/ReasonForge-MCP-Serverupdated this year
A deterministic tool-calling framework for small LLMs. Integrates SymPy and sandboxed Python to achieve better accuracy/latency without massive parameter counts.
Apache-2.0 -
zest-langgithub.com/r2jitu/zest-langupdated this year
Token-efficient programming language for LLM tool calls; transpiles to Python; ships with MCP server
★ 1 -
nimai-mcpgithub.com/balagopalaji/nimaiupdated this year
Nimai MCP server — exposes nimai_spec, nimai_review, and nimai_validate as tools. No internal LLM calls.
npm MIT -
mcp-turboquantgithub.com/ShipItAndPray/mcp-turboquantupdated this year
MCP server for LLM quantization. Compress any model to GGUF/GPTQ/AWQ in one tool call. First MCP server for model compression.
MIT ★ 5 -
evalflow-mcpgithub.com/arko-14/evalflow-mcpupdated this year
MCP server for the EvalFlow API — exposes EvalFlow (LLM prompt evaluation) as tools any MCP-capable CLI or AI agent can call.
npm MIT -
customclaw-mcpgithub.com/onlythebestswfl-ops/customclaw-mcpupdated this year
MCP server for the CustomClaw registry. 37 vetted utilities for LLM and agent code (rate limiters, JSON repair, response caches, retry wrappers) installable as native tool calls from Claude Desktop, Claude Code, Cursor, Windsurf, and any MCP host.
npm MIT -
rust-docs-mcp-servergithub.com/Govcraft/rust-docs-mcp-serverupdated this year
🦀 Prevents outdated Rust code suggestions from AI assistants. This MCP server fetches current crate docs, uses embeddings/LLMs, and provides accurate context via a tool call.
go MIT ★ 301 -
MCP server for Gate402: pay-per-call agent APIs over x402. Open-model LLM inference, Base trading data, SEC filings and web scraping. 20 tools, free tier, no signup.
npm MIT ★ 1 -
fun-strapi-ai-plugingithub.com/PaulBratslavsky/fun-strapi-ai-pluginupdated this year
Includes LLM chat, tool calling, and MCP server.
-
Voice control for edge devices: on-device agent (listen → LLM tool-calling → device action → speak). Swappable cloud/local backends. By embedai.com.cn
MIT -
azure-vm-teams-botgithub.com/ferryiswono/azure-vm-teams-botupdated this year
Manage Azure VMs from Microsoft Teams in plain language — MCP server + Bot Framework middleware with LLM tool calling
MIT -
MCP server wrapping knowledge-gun. Exposes paste-ready knowledge-graph context bundles as tools any MCP-aware LLM client can call.
MIT -
orbit-mcp-servergithub.com/AyoubProjects28/orbit-mcp-serverupdated this year
MCP server exposing tools the LLM orchestrator can call during chat.
-
Context compression for AI coding agents: compresses tool output before it enters the model, dedups repeats to 13-token refs. Claude Code, Cursor, Codex, Kiro, Zed, any MCP client. Rust, zero LLM calls.
cargo Apache-2.0 ★ 677 -
Shared FalkorDB context graph for coding agents: a ~200-token map before the first tool call, memory across sessions, edit claims between agents. 144 Claude Code runs: −25% tokens, −29% tool calls. No LLM calls, no Docker.
MIT -
Cheapest-LLM Router — a zero-dependency MCP server that routes any prompt to the cheapest model that can actually serve it: free models first, then the lowest-cost paid fallback. Returns USD/CNY cost, a fallback chain and plain-English reasoning. 4 tools, no npm install, calls no model itself.
MIT -
See what your AI edits affect. Structural checks for AI code changes. Local call/import graphs and 28 MCP tools for impact, callers, and duplication. No LLM in the checks.
MIT -
hpml-assetopsbench-smart-grid-mcpgithub.com/eggrollofchaos/hpml-assetopsbench-smart-grid-mcpupdated this year
SmartGridBench: a smart-grid transformer-maintenance benchmark for tool-using LLM agents, built on IBM AssetOpsBench with MCP/direct-call tool surfaces.
MIT -
hpml-assetopsbench-smart-grid-mcpgithub.com/HPML6998-S26-Team13/hpml-assetopsbench-smart-grid-mcpupdated this year
SmartGridBench: a smart-grid transformer-maintenance benchmark for tool-using LLM agents, built on IBM AssetOpsBench with MCP/direct-call tool surfaces.
MIT ★ 1 -
An interactive Model Context Protocol (MCP) server & client implementation in TypeScript. Features dynamic tool calling, custom prompts, resource templates, and bi-directional LLM sampling integrated with Groq LLMs and the Vercel AI SDK.
★ 1 -
A deterministic policy firewall for AI agent tool calls. YAML rules, no LLM calls, no risk scores. CLI + MCP server (advisory) + @enforce() decorator (unbypassable). MIT, 46 tests, zero network dependency.
pypi MIT ★ 1 -
Security & Observability for AI Agents — audit every tool call, enforce allow/block policies, catch prompt injection and data leaks. Local-first; works with Claude Code, Codex, Copilot CLI, OpenClaw, LangChain, and all major LLM APIs.
npm NOASSERTION ★ 15 -
Fully local AI stack in Docker Compose — Ollama LLM inference, SearXNG private metasearch, Firecrawl scraping + real browser automation, and Langfuse observability behind one authenticated gateway with an MCP tool interface. No cloud calls, your data stays on your hardware.
MIT