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.
RAG Codebase
Code-aware RAG over a repository: AST-boundary chunking (tree-sitter), hybrid dense+BM25 retrieval with RRF & reranking, a code graph (imports/calls + PageRank), grounded citation…
Alpha
Korean crypto × AI vertical media + community — alpha.moss.land. Channel stance, AI briefs, RAG, 8 AI personas with track records, 12-tool MCP server.
Claude Workspace
🧠 Production-ready Claude Code workspace — 13 skills, 8 agents, 11 slash commands & 10 hooks for React/TypeScript frontend teams. One-command install.
Metis
Local-first AI dev agent for your terminal — provider-agnostic, cost-aware, self-improving. Apache-2.0.
Meridian
隐私优先的 AI 桌面客户端——编程 agent、多模型接入、内置输入法。| Privacy-first AI desktop client with coding agent, multi-provider support, and a built-in input method
Emcp
Enhanced MCP — typed, pixel-accurate, cross-server context for LLMs
Agent Gateway
Standardized gateway for AI agent discovery, custom contracts, and secure auditing with Guardrails
AI Engineer
AI engineering portfolio with Next.js demos for LLMs, RAG, vector search, embeddings, agents, MCP, guardrails, observability, prompt engineering, and token economics.
Librarian
A personal knowledge layer for coding agents. Ingest, compile, search, lint -- Karpathy's LLM Knowledge Base pattern as a real tool.
Memforge
Let AI coding assistants remember your team's knowledge across sessions
Hello World
Local-first AI workspace for Web, Windows Desktop, and Mobile with multi-provider chat, RAG, Web search, MCP/plugin tooling, and Desktop-only sandboxed code execution.
Allan MCP Memory Code
Store persistent knowledge graphs for AI coding assistants using a local Docker setup with hybrid search and entity extraction.
Polyglot
Write in your native language, Claude Code responds in English. Local-GPU translation + multi-layer anti-bypass enforcement. ~18% tokens saved on Polish, ~43% on CJK.
AIDaemon
Lightweight AI framework providing access to mcps, skills, scheduled jobs, persistent memory etc.
Context Graph
Decision-aware knowledge graph on LightRAG: extends (head, relation, tail) triples into contextual quadruples (h, r, t, rc) that capture decision lineage — who approved what, why,…
Awesome AI Application Engineer
Practical AI application engineering roadmap: LLM basics, Prompt Engineering, RAG, Agents, MCP, evaluation, production, tutorials, checklists, and templates.
Headroom
More human oversight can make an AI agent less safe. Headroom is a human-in-the-loop firewall for coding agents that measures when to trust the human: oversight as resource alloca…
Dotbot
agent harness — multi-provider, streaming, 47 tools, autonomous tasks
Hydra
Real-time AI threat interception. Blocks CVE-mapped attacks, poisoned configs, and destructive terminal commands before execution.
Hermes Prime
Drift-prevention session-init convention card for fresh Claude Code sessions. Injects a self-contained card so a new session opens with its grounding triggers, calibration rules,…
AI Toolkit
Production-ready AI coding skills and agents for Claude Code, OpenCode, and Pi. Covers TDD, .NET, Python, Rust, edge AI, and more.
BLADE
Your AI that watches, learns, and works — parallel agents, screen memory, God Mode, MCP, voice. Built with Tauri + Rust. Free, open source.
Context Pack
Task-specific repo context bundler for Claude Code & OpenAI Codex
Skills
Open-source GTM playbook as skills for Claude Code, Cursor, and other LLM clients | skills.reachrobin.com
Prism
PRISM: Persistent Reasoning & Intelligent Self-evolving Multi-agent Framework — 持久推理与智能自进化多智能体框架
Finance Skills
AI agent financial skill — real fundamentals, deterministic math, fail-closed. Stop agents inventing stock numbers.
Forge
Forge – The AI Execution Layer
Gateway
统一接入多 MCP Server 的智能体网关：stdio/SSE/HTTP/inprocess 四传输 · 语义工具路由 · 网关内 Agent · SSE 流式 · 多租户白名单 · 断线自愈 | FastAPI · 156 tests · CI · Docker · Live demo
AI Agent Engineering
AI Agent 系统工程深读笔记 | 从 LLM→Agent→Orchestrator 的架构演进、关键技术与工程实现 | Session Management · Context Engineering · Tool Registration · Harness Engineering
Deepseek Harness
Local-first, safety-gated CLI and MCP harness for bounded DeepSeek batch and corpus workloads.
ScarabHive
Self-hosted framework for LLM agents, built around plugins: agents defined in YAML, run in the browser, from the CLI or through an OpenAI-compatible API.
Trimum
trimum — Linux AI Agent runtime (harness): daemon + pluggable agents + unified tool gateway + sandbox + tamper-evident audit
Momputer
A mom for you, your computer and your AIs. Turns off what AI agents left running, frees up disk space, and checks your drives, memory, sound and Windows settings. For Windows.
Ink
An open-source AI assistant and agent harness for the terminal — extensible via skills, subagents, and MCP; works with Anthropic, OpenAI, GLM, DeepSeek, and Kimi.
Agents
Personal AI agents built on GoFr 1.58 — LLM + MCP agent loop, SSE streaming, RAG. Runs keyless via a claude-CLI shim; default provider Groq.
Agent Shield
为 LLM Agent 提供安全防护与审计框架，防御 MCP 投毒攻击，支持 HITL (Human-in-the-loop) 与审计调度。
Workbuddy Bridge
Use a WorkBuddy desktop account as an OpenAI- and Anthropic-compatible API for Claude Code, Codex, and any coding agent.
AI SDK
Long-term memory for agents built with the Vercel AI SDK (ai 5, 6 and 7): the Mnemoverse MCP memory tools, run in process.
Hippocampus
Shared, curated memory for a party of AI agents — an Obsidian vault run like a TTRPG campaign wiki. Agents write episodes; a nightly sleep consolidates them into canon.
Evoflow
EvoFlow: 三层进化栈的自进化 LLM Agent 框架 / Self-evolving LLM agent framework (three-layer evolution stack)
Nis
Self-hosted AI agent workbench - long-term memory, knowledge base, full toolchain. Desktop + mobile web + QQ/WeChat, bring your own model. With you, I can do anything. Run by me,…
Claunity
AI assistant inside the Unity Editor: chat with your scene, write scripts, fix errors, test and plan projects. Local Python backend, bring your own Claude API key. Unofficial, MIT.
Spine
Spine — доменный харнесс solution-архитектора: spine-инварианты, ADR, рубрики с LLM-судьёй, скиллы/плагины, субагенты, губернанс · A domain agent harness for solution architects (…
ToolkitArchive
Verified archive of AI models, coding agents, free AI access, skills & MCPs, benchmarks and $0 alternatives — updated Sept 2026
Verdict Core
Control plane for AI coding agents: decides which models may run each task, recovers from rate limits and timeouts, sends work to a separate reviewer model, and writes verifiable…
AI Agent
生产级 Agent Engineering 知识库与工程体系，覆盖 LangGraph、RAG、MCP、Multi-Agent、Runtime、Evaluation、Security、SRE 与 Skills。
Erp AI Agent
Open-source enterprise AI agent: LLM tool-calling engine, permissions, RAG and a Python analytics engine, with a working ERPNext connector. Architecture guide: noviz.in/how-to-bui…
Checkpoint
The release gate for AI agents — run your real agent N times against stateful twins and ship only when the confidence interval says so.
Igor
A minimal, dependency-free coding agent written in C
Iris
LLM-first, source-agnostic messaging system. Normalize messages from Telegram, SMS, WhatsApp, Email, and more into a unified API. Self-hostable. MIT.