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. 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.
-
litellm-mcpgithub.com/wklee610/litellm-mcpupdated this year
MCP server for LiteLLM
MIT ★ 3 -
LiteLLM Vector Store MCP Integration
★ 1 -
LiteLLM-MCP-Servergithub.com/ArtemisAI/LiteLLM-MCP-Serverupdated this year
MCP Server for LiteLLM API integration
npm MIT ★ 2 -
@semidark/mcp-litellm-searxnggithub.com/semidark/mcp-litellm-searxngupdated this year
MCP server for LiteLLM proxy SearXNG search endpoints
npm MIT -
litellm-mcpgithub.com/shin-bot-litellm/litellm-agent-mcpupdated this year
MCP server giving AI agents access to 100+ LLMs through LiteLLM
pypi MIT -
MCP server exposing multi-provider LLM completion via LiteLLM — cross-provider fallback, pre-call budget enforcement, per-request cost attribution.
MIT -
googleADK_vth_LiteLLMgithub.com/AnithaKarre/googleADK_vth_LiteLLMupdated this year
A Figma design-to-code agent built on Google ADK and LiteLLM — extracts design tokens (text, colors, layout) from Figma via a custom MCP server and generates JSX-ready code. Supports multi-provider LLM routing and modular prompt templates.
-
@atriumn/tariffgithub.com/atriumn/tariffupdated this year
LLM Pricing Oracle — MCP server exposing real model pricing data from LiteLLM
npm MIT -
LLM Pricing Oracle — MCP server with real pricing data from LiteLLM
npm MIT ★ 1 -
OAuth2-protected MCP server for on-premises Microsoft Exchange via EWS, with Keycloak and LiteLLM integration.
MIT -
MCP-palvelin, joka täydentää LiteLLM:ää: mallin valinta tehtävätyypin mukaan, kulutuksen seuranta ja eval-vertailu tekoälyagenteille.
MIT -
Self-hosted family voice assistant: own ESP32-C3 firmware (no OTA) + Python server, LiteLLM models, MCP tools. MIT.
MIT -
hybrid-model-switcher-mcp-for-Claude-Desktopgithub.com/berry1200/hybrid-model-switcher-mcp-for-Claude-Desktopupdated this year
MCP routing proxy for Claude Desktop — local Ollama model switching via LiteLLM
MIT ★ 3 -
mcp-uber-eats-apigithub.com/Nathan7128/mcp-uber-eats-apiupdated this year
Ce projet combine un serveur MCP exposant des outils connectés à l'API Uber Eats Marketplace avec une boucle agentique LiteLLM.
-
MCP server for extracting entities from text chunks and creating entity graphs in Neo4j. Supports 100+ LLM providers via LiteLLM.
pypi MIT -
VFA — Visual & Functional Auditor: auditoria web autonoma (visual + funcional) expuesta como servidor MCP, con Playwright local y motor universal LiteLLM.
pypi LicenseRef-Proprietary -
MCP speech-to-text server: FastAPI + fastapi-mcp, FFmpeg preprocessing, and policy-based routing across OpenAI and Groq via LiteLLM.
-
Agentic AI middleware: a circuit breaker for autonomous agents — FastAPI/LiteLLM reverse proxy that tracks token spend, enforces budgets, and aborts runaway loops. TypeScript MCP server included.
-
Open-source agentic AI for revenue intelligence — 6 specialized agents (extract, resolve, compute, detect, explain) with BYOM via LiteLLM. CLI + MCP server. pip install vesh-agents
Apache-2.0 -
levisnkyyyy-images-mcpgithub.com/yourusername/images-mcpupdated this year
Model Context Protocol server for AI image and video generation using LiteLLM and fal.ai
npm MIT -
Self-hosted Draw.io editor extended with an AI-powered conversational sidebar — generate, modify, and refine diagrams using natural language via MCP tools and LiteLLM
-
Local agent-to-agent messaging over MCP for Codex CLI, Claude Code, and optional LiteLLM API models. SQLite mailboxes, resumable sessions, and two-way conversations.
MIT -
agentic-test-generatorgithub.com/Hrushitha12/agentic-test-generatorupdated this year
5-agent AI system using CrewAI, MCP, and A2A Protocol to autonomously generate API test suites from OpenAPI specs — GPT-4o/Claude Sonnet via LiteLLM with LangSmith tracing
-
ИИ-сотрудник для коробочного Битрикс24 (закрытый контур, LLM через LiteLLM): чат-бот, таймлайн, письма и файлы на Диске, CRM и смарт-процессы, задачи, участие в бизнес-процессах + MCP-сервер.
MIT -
Guardrino by HisnLabs: the AI guardrail for Arabic, its dialects, Arabizi and English. Stops prompt injection and jailbreaks, masks Gulf personal data. OpenAI-compatible proxy, LiteLLM plugin, MCP security proxy. Built in Dubai.
Apache-2.0 ★ 1 -
A local, containerized AI workbench. Open WebUI over LiteLLM, an MCP tool rack where every tool is an OCI container, and Phoenix tracing across the lot. Built as a starting point to fork, not a product to adopt.
Apache-2.0 ★ 2 -
MCP server that delegates Claude Code subagents to local models (LiteLLM/llama.cpp/Ollama), DeepSeek, AWS Bedrock, or any OpenAI/Anthropic-compatible backend — without losing the Claude Code orchestrator session.
MIT ★ 19 -
Which AI model should you use, and what will it cost? Every price change in LiteLLM's catalogue since 2023, Arena ratings, a competing risks model, an MCP plug-in and an Agent Skill. Rebuilt daily.
pypi MIT -
ollama-mcp-chat-cligithub.com/Shahrukh19S/ollama-mcp-chat-cliupdated this year
A terminal MCP chat agent on a free local model (Ollama) with Groq/GitHub fallbacks - implements an MCP Server (tools, resources, prompts) and an MCP client with an agentic tool-calling loop via LiteLLM
MIT -
ValtAG (Valt for Agents) — the ask-first credential gate for AI agents: agents request, humans approve, leases expire, everything audited. Self-hosted MCP-native vault. Config-only integration with nginx, Caddy, Traefik, Envoy, APISIX, Higress and LiteLLM.
Apache-2.0 -
Private, self-hosted AI platform that runs on one laptop (no cloud, no GPU, no API keys): agentic RAG with citations, LangGraph + MCP tools, LiteLLM routing with an evaluation gate, MLOps (Dagster, MLflow, BentoML), OpenTelemetry tracing, SSO and Argo CD GitOps on k3s with Ollama.
MIT ★ 9 -
LiteLLM proxy hook and MCP server for chat bots in groups: three-scope memory (person, group, shared) whose scope is signed by the gateway, never chosen by the model; automatic recall; document routing; and an output filter that turns a tool call written as text into an honest admission.
MIT -
A local CLI and MCP server for talking to configured AI models through LiteLLM.
pypi MIT License -
mock-oauth2-mcp-servergithub.com/BerriAI/mock-oauth2-mcp-serverupdated this year
A mock OAuth2 + MCP (Model Context Protocol) server for testing client_credentials flows. Useful for E2E testing LiteLLM proxy MCP OAuth2 M2M authentication.
★ 3 -
Python CLI and MCP server for Higgsfield AI image and video generation. Supports AI agent workflows, media uploads, asynchronous requests, polling, cancellation, cost estimates, optional LiteLLM cost tracking, and downloads.
BSD-2-Clause ★ 1 -
mcp-resilient-researchgithub.com/stroccoli/mcp-resilient-researchupdated this year
Autonomous multi-provider web research pipeline exposed as an MCP server — LangGraph + FastMCP + LiteLLM, deployed on Azure Container Apps
MIT -
MCP server for text-to-speech through any OpenAI-compatible API (OpenAI, OpenRouter, LiteLLM, local). Saves audio to the paths the agent asks for.
npm MIT -
LLM agent orchestration for scientific workflows. Python 3.12, LiteLLM, BoTorch, ASE, Streamlit UI, MCP server and client.
-
多平台热点聚合 + AI 分析 + 多渠道推送 + MCP 查询引擎(基于 TrendRadar v6.5.0)。11 平台热榜 + RSS,时段化调度,9 类通知渠道,LiteLLM 分析,FastMCP 26 工具。 | Multi-platform trending aggregator + AI analysis + multi-channel push + MCP query engine (based on TrendRadar v6.5.0).
GPL-3.0 -
Persistent memory for AI coding agents: decisions, rules, API contracts, and system topology served over MCP to Claude Code, Cursor, Codex, and Gemini CLI. Powers AI pull request review, issue triage, and a local-first code graph that syncs to the cloud. Self-hosted, 100+ models via LiteLLM.
MIT ★ 35 -
local-agent-mcpgithub.com/RGGH/local-agent-mcpupdated this year
Smol Qwen MCP + Agent Tool Calling Example
-
consult-mcp-servergithub.com/irwin-r/consult-mcp-serverupdated this year
Multi-model LLM panel consultation engine with an MCP adapter.
pypi MIT ★ 2 -
GISgithub.com/devdattatalele/GISupdated this year
GIS — Autonomous GitHub Issue Solver. RAG-powered CLI agent + MCP server. pip install gis-cli
MIT ★ 2 -
Local-first personal AI efficiency layer — MCP server for prompt improvement, template library, usage tracking, and scoped memory.
MIT ★ 1 -
mcp-second-opinionupdated this year
Let your MCP-aware agent consult rival LLMs (OpenAI, Gemini, Anthropic, Grok) mid-conversation.
pypi MIT -
One-click knowledge system for documents, internal bots, and AI agents
MIT ★ 51 -
A CLI, Agent Skill, and MCP server for consulting AI models with persistent conversations and image support.
MIT ★ 13 -
Vendored fork of stephanschielke/open-webui-mcp-server packaged for container deployment behind an authenticating MCP gateway.
MIT ★ 2 -
Space Summit 2026 workshop: end-to-end satellite telemetry with Flink, Iceberg, Trino, Superset, and an MCP server.
★ 1 -
chunktunergithub.com/shantanu-deshmukh/chunktunerupdated this year
Benchmark chunking strategies for your RAG corpus. Get a recommended config. CLI, Python library, and MCP server.
pypi MIT ★ 2 -
Multi-Agent-Mcpgithub.com/R-Mabasha/Multi-Agent-Mcpupdated this year
A Model-Agnostic, Git-Sandboxed Multi-Agent Nexus MCP Server.
MIT ★ 3 -
Flight & Hotel MCP Servers + A2A Agents
Apache-2.0 -
MCP-native LLM councils for debates, juries, blind panels, voting, refinement, and custom multi-model deliberation.
npm NOASSERTION ★ 1 -
@joezm/seed-vizupdated this year
Vision analysis CLI + MCP server backed by Seed 2.0 via Volcano Ark or any OpenAI-compatible endpoint
npm MIT -
Measures how often LLMs fail to return valid structured output, and what fixes it. MCP validator: uvx typecast-mcp
pypi MIT ★ 1 -
career-scout-mcpgithub.com/stestojadinovic/career-scout-mcpupdated this year
Production-grade MCP server demonstrating LLM-based job posting scoring with provider abstraction, strict input validation, and minimal trust surface.
MIT -
MCP server that documents a repo for you — analyzes the stack, screenshots the deployment, generates charts, drafts a README and launch posts
-
Tool-calling SQL agent with read-only guardrails, an MCP server and an evaluation harness (bootstrap CIs, ablations, prompt-injection test)
-
Give your AI agents an evaluation tool - provider-agnostic LLM eval runner with an MCP server for Claude, structured output validation, datasets, evaluators and honest cost metrics
pypi MIT -
Production-grade Strands appointment scheduling agent: real CMS provider search, FastAPI scheduling service, MCP tools, session + three-layer memory, PHI-by-reference, deterministic form filling.