MCP servers tagged llm
10K live MCP servers tagged "llm", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with llm.
Rlm MCP
MCP server implementing Recursive Language Model pattern for processing arbitrarily long contexts. Enables Claude Code to work with 1M+ character documents through session-based c…
Cairn
Shared memory for AI coding agents.
Mycelium
Resident cross-tool skill brain for AI coding agents — one shared memory across Claude Code, Codex & more over MCP.
Engram Vscode Plugin
VS Code plugin for Engram that adds persistent memory lifecycle hooks to GitHub Copilot sessions.
AIttache
Give your LLM eyes, not hands. An MCP server for people who don't want a golden retriever with sudo
SDK Typescript
ZiroAgent SDK — production-safe TypeScript agent runtime: durable execution, cost guardrails, replayable traces, MCP-native, sovereign-ready. https://ziroagent.com
Json Contracts Studio
Local web Studio for testing natural-language-to-JSON workflows with JSON Contracts.
UnboxAPI OpenSpec
UnboxAPI OpenSpec — SemanticMap schema (Apache-2.0). Open spec only; reference implementation and tooling remain proprietary in UnboxAPI.
SIMBIOLOGY MCP
An MCP server providing AI assistants with access to MATLAB SimBiology for modelling, simulating, and analyzing biological systems
Agentic
An agent is just a loop. This free book and site prove it by building one from scratch with Python and Ollama: tool calling, RAG over a real PDF, layered guardrails, and a live MC…
PinkRooster
Tool collections for AI agents on the Microsoft Agent Framework: tools, prompt text and live state in one class, plus an agent builder, MCP and run events. .NET 10.
Agent Opt
Optimize Claude Code, Codex, Cursor & Antigravity with smarter context, fewer tokens, faster coding and automatic verification.
Probe
Probe any MCP (Model Context Protocol) stdio server in one command — handshake check, full tool listing, and a safe smoke-call per tool with CI-friendly exit codes. Pure Python st…
Runpost
The HTTP client you and your AI agent share: a Postman-like desktop app (Rust + Tauri) with a built-in MCP server for Claude Code, Codex and other agents. One binary, no account,…
Ragforge
Hybrid RAG toolkit — BM25 + vectors with RRF, from-scratch HNSW, guardrails, verifiable citations, agentic retrieval and an MCP server. Zero dependencies.
Neuralis
Self-hosted AI workforce: a multi-agent, multiplayer workspace where agents and your team share projects, files and tools — built around one package protocol.
Saturn App
Run AI agents on your own computer: a desktop agent app, a local runtime and an IDE in one download. Use Claude, OpenAI, Gemini, DeepSeek or local models, with permission rules, a…
Project Intelligence
Local-first, vendor-agnostic AI project orchestration and governance framework with lifecycle gates, quality validation, memory, and MCP integration.
MrKyaw
Principal AI Engineer & Forward-Deployed AI Architect | Enterprise & Solution Architecture | Agentic AI, Multi-Agent Systems, RAG & MCP | MLOps/LLMOps, Evaluation, Observability &…
Termux MCP
Termux MCP server for Android - let AI agents control your phone via shell commands, file I/O and package management. 安卓 Termux 的 MCP 插件，实现安卓端 AI Agent 对手机的掌控，readme 附提权教程
Smart Context Packer
📦 Ultra-polyglot AI context packer & 15 MCP tools for LLM coding assistants (Cursor, Claude, Copilot, Codex). Zero Python/DB bloat.
Contextetech
Open-source European platform for verified AI context: prompts, guardrails, datasets and evals, each with its source and EU compliance. contextetech.com/en/
Cross Session MCP
MCP server that hands off coding sessions across Claude Code, Codex, Cursor, Copilot, OpenCode and Antigravity. Gemini reads the old transcript so your agent doesn't have to: ~25x…
Tierdecay
Per-repo learning layer for AI coding model routers: learns which task classes can safely run on a cheaper tier. Native for Claude Code, Codex, Antigravity · MCP server · zero dep…
Codex App Server MCP
MCP server that lets Claude Code drive OpenAI Codex via codex app-server — persistent threads, async turns, resume, interrupts.
Redrob Cowork
Redrob Cowork is a free, open-source desktop and MCP app for doing work with AI agents on your own files.
A11y Computer Use
a11y-computer-use: accessibility-first computer use for AI agents. Pruned a11y-tree snapshots, element refs instead of pixel coordinates, a safety layer, an MCP server and CLI. ma…
Librechat Wren Demo
One-command "ChatGPT for your data" demo: LibreChat + Wren AI MCP text-to-SQL over a mock retail Postgres warehouse, with SearXNG web search and chart artifacts. Runs in Docker Co…
LLM MCP Fastapi
FastAPI + MCP server for LLM inference — chat API with tool-calling, streaming responses, and model-context-protocol support.
Quarkus Langchain4j AI
Java LLM chat app on Quarkus + LangChain4j with an MCP server module, local models via Ollama (Claude optional), and a React UI
Ctrlyoke
One workflow engine for every CLI coding agent
Hybrid Rag
Production-grade hybrid RAG in TypeScript — vector + BM25 + reranker across 14 vector stores (Qdrant, Pinecone, Weaviate, pgvector, LanceDB & more) behind one adapter API. Benchma…
Flow
A disciplined five-phase loop (Discuss → Plan → Execute → Verify → Ship) for Claude Code & GitHub Copilot. Heavy work runs in fresh-context subagents, verify gates ship, and token…
Pm Assistant
Local-first AI copilot for project managers: MCP tools across Jira, GitHub, Slack and more, approval-gated writes, natural-language rules engine.
Orcheus AI
Desktop AI coding assistant built with Electron, Node.js/Express, Supabase and Flowise, featuring MCP file tools and SSE streaming.
Argus
Framework-free, horizontally-autoscaled multi-agent deep-research engine. Own the loop, not the framework.
Neuron Js
AI-friendly TypeScript rules engine for serializable JSON business rules and deterministic workflow decisions — LLM guardrails, agent-readable docs (llms.txt), MCP-ready.
Anti Slop
KB-grounded AI consultation with citation discipline. Skills + gates + schema. Methodology at anti-slop.dev.
Almanac
Compile domain-specific, freshness-aware knowledge surfaces as CLI tools, MCP servers, and Claude Code Skills.
Memoryos
Episodic + semantic memory backend for long-running AI agents
Econoclast
An adversarial AI referee for empirical economics: finds p-hacking, cherry-picking and specification search, then re-runs the result.
Omnex
Universal retrieval. Structure-aware. Token-efficient.
Claude Mem
Zero-dependency session recall for Claude Code — instant memory across compactions, ~50 tokens per prompt
Synapse
Open-source MCP server that gives Claude persistent memory. Gemma compresses your codebase into a searchable vault — no more re-explaining your stack every session.
Claude Code
Explore Claude Code CLI source code for TypeScript, Bun, and Ink, with tools, commands, and services for AI coding automation
Claude Code
Use Claude Code CLI source to inspect, edit, and automate terminal-based software engineering workflows with AI assistance
Memilier
Hemilier 桌面智能体：对话 + 工具调用 + 文件/命令操作 + 联网搜索，由本地 Ollama 或云端 DeepSeek/OpenAI 兼容模型驱动。Electron + React + TypeScript。
AgentForge
AI software engineering platform that turns natural-language requests into working, tested, documented codebases. An 8-agent LangGraph pipeline (analysis, planning, architecture,…
Weather Project Copy2
AI-powered weather agent that uses the Model Context Protocol (MCP) to connect OpenRouter LLMs with real-time OpenWeatherMap data, exposed via a FastAPI MCP server and a Streamlit…
AgentRank
The trust and reputation layer for AI agents. Protocols like MCP and A2A tell you how agents connect. AgentRank tells you whether you should connect.