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 descriptions, licences and links are read from each package's own registry (npm, PyPI, crates.io, NuGet, RubyGems; Go and Maven packages are only checked to exist), and packages the registry no longer has are removed. 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.
-
llm-code-cache-mcpgithub.com/TroublingStone/llm-code-cache-mcpupdated this year
Code Base Intelligence MCP with Graph RAG
-
MCP server for running source code security reviews with an LLM
Apache-2.0 -
rlm-mcpgithub.com/adrianwedd/rlm-mcpupdated this year
MCP server implementing Recursive Language Model pattern for processing arbitrarily long contexts. Enables Claude Code to work with 1M+ character documents through session-based chunking, BM25 search, and artifact provenance tracking.
★ 1 -
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 -
deep-mcpgithub.com/Todmy/deep-mcpupdated this year
Recursive Language Models (RLM) — CLI + MCP server with code-execute / recursive-delegation loop
-
rlm-mcpgithub.com/eesb99/rlm-mcpupdated this year
MCP server for Recursive Language Models (RLM) - verified code execution with LLM reasoning
MIT ★ 2 -
Recursive Language Models for Claude Code - Infinite memory solution inspired by MIT CSAIL paper
pypi MIT ★ 39 -
memcpgithub.com/maydali28/memcpupdated this year
A persistent memory MCP server for Claude Code - Recursive Language Model integration for Claude Code inspired by MIT CSAIL paper
pypi MIT ★ 18 -
MCP + REST server exposing fast search over Linux kernel mailing list archives (lore.kernel.org) to LLM developer tools. Python 3.14 + Rust (PyO3) with custom three-tier index: columnar metadata, trigram code search, BM25 prose.
pypi MIT ★ 5 -
Repair the JSON language models actually emit: code fences, trailing commas, smart quotes, unquoted keys, prose noise. MCP server + skill. Offline, deterministic, tested.
Apache-2.0 -
Run any AWS Lambda function as a Large Language Model (LLM) tool without code changes using Anthropic's Model Context Protocol (MCP).
MIT ★ 110 -
An open-source Python MCP (Model-Command Protocol) server designed to enable Large Language Models (LLMs) to execute FFmpeg commands, receive execution results, and validate commands against FFmpeg source code.
MIT -
Local data workbench a local language model drives over MCP: ingest, versioned cleaning, quality rules, charts, dashboards as code and quick models. Part of the Hoard family.
MIT -
A Model Context Protocol (MCP) server that provides file system context to Large Language Models (LLMs). This server enables LLMs to read, search, and analyze code files with advanced caching and real-time file watching capabilities.
MIT ★ 40 -
The Meta Ads MCP Server is an implementation of the Model Context Protocol (MCP) that enables Large Language Models (LLMs), such as Claude Code, to interact directly with the Meta Ads API.
-
rlmgithub.com/SaschaOnTour/rlmupdated this year
Context broker for AI coding agents (CLI + MCP). Queries codebases via progressive disclosure instead of feeding entire files into context — reducing token waste and context rot. AST-aware retrieval, surgical editing with syntax guard, semantic search. Inspired by Recursive Language Models research (MIT/Stanford, 2025). Built in Rust.
NOASSERTION ★ 2 -
code-index-mcpgithub.com/johnhuang316/code-index-mcpupdated this year
A Model Context Protocol (MCP) server that helps large language models index, search, and analyze code repositories with minimal setup
MIT ★ 1,005 -
Little Helper bridges your Godot 4 game engine with modern Large Language Models and autonomous coding agents
MIT -
faust-mcp-testgithub.com/jessesep/faust-mcp-testupdated this year
Faust DSP Language Model Context Protocol Integration - Comprehensive documentation and tools for Claude AI to write, test, and debug Faust code
MIT ★ 2 -
A Model Context Protocol (MCP) server that builds rich code graphs to provide deep contextual understanding of TypeScript codebases to Large Language Models.
npm MIT ★ 18 -
🤖 Enable AI code generation using the Code Execution MCP Server, integrating large language models with the Model Context Protocol for seamless tool use.
★ 1 -
Intelligent code editing powered by tree-sitter-gap, GAP code running, a help system, language model tools and MCP support
MIT ★ 4 -
Fast local code search, source reads, and smart block-clamped context for AI coding agents. VS Code Native Language Model Tools & MCP server backed by a shared Rust trigram cache (Linux, Windows, WSL).
MIT -
censuschatgithub.com/smach/censuschatupdated this year
censuschat aims to let you ask questions about U.S. Census data in natural language. More specifically, a large language model takes queries about Census data and translates them into tidycensus R code. It uses the ellmer R package under the hood.
NOASSERTION ★ 14 -
llm-sca-toolinggithub.com/grammy-jiang/llm-sca-toolingupdated this year
Evidence-first, LLM-augmented static code analysis: graph indexing, MCP server, audit workflows
pypi MIT -
local-llm-delegategithub.com/faizan007jr/local-llm-delegateupdated this year
Claude Code plugin that delegates simple, well-scoped tasks to a locally running LLM (Ollama, LM Studio, llama.cpp, vLLM) via MCP
MIT ★ 1 -
Share code with LLMs via Model Context Protocol or clipboard. Rule-based customization enables easy switching between different tasks (like code review and documentation). Includes smart code outlining.
pypi Apache-2.0 ★ 308 -
Universal LLM router for AI coding tools. Works with Claude Code, Cursor, Codex, Gemini CLI, Copilot and more. Free-first fallback chain cuts costs 35-80%.
npm MIT ★ 130 -
Zero-dependency MCP server and CLI that puts Codex, Claude Code, Grok, Cursor, Antigravity, Pi, OpenAI and local LLMs behind one interface
Apache-2.0 -
llm-council-mcpgithub.com/JeremyGracey-AI/llm-council-mcpupdated this year
Multi-model LLM Council (Karpathy-style 3-stage deliberation over OpenRouter), exposed as an MCP server for Claude Code.
MIT ★ 1 -
mcp-code-reviewgithub.com/mauriziomocci/mcp-code-reviewupdated this year
MCP server for LLM-powered code review — GitHub PRs, GitLab MRs, local diffs
pypi MIT -
Run Claude Code on any LLM: a Cloudflare Workers AI gateway and LLM proxy that speaks the Anthropic Messages API and routes to Workers AI, OpenAI, Gemini, DeepSeek or Anthropic, with a prompt-injection session guard, red-team harness and MCP server.
MIT -
llm-council-mcp-servergithub.com/salutaris91/llm-council-mcp-serverupdated this year
Local MCP server running Andrej Karpathy's 3-stage LLM Council via OpenRouter — a multi-model "second opinion" for Claude Code, Codex CLI & Antigravity. By Alexander Deja · anderzlabs.de
pypi MIT ★ 9 -
llm-api-docs-ledgergithub.com/amazadfar/llm-api-docs-ledgerupdated this year
Local-first retrieval ledger for LLM provider docs, with deterministic snapshots, SQLite/FTS5 search, Markdown exports, API endpoints, and MCP tools for coding agents.
MIT -
Deterministic code-graph (GraphRAG) over your repo for LLM agents — local-first, git-native, zero-infra, served via MCP. Python, TS/JS, Rust, Go, Java, C#.
npm Apache-2.0 ★ 14 -
See every hop of a request through your local LLM system: a tracing server, a live flow map, a waterfall per request, and an MCP server for Claude Code and Cursor. One static Go binary.
Apache-2.0 ★ 1 -
A real-time semantic codebase knowledge graph powered by a local LLM for macOS, delivering clean architectural context to AI coding agents through native MCP.
npm MIT ★ 1 -
code-review-crewgithub.com/siddlpskmi/code-review-crewupdated this year
Autonomous multi-agent AI code reviewer with custom MCP server, agentic RAG, and LLM-as-judge evals
-
alt-llm-plannergithub.com/VikAnalytics/alt-llm-plannerupdated this year
MCP tool for Claude Code & Cursor: offload architectural planning Q&A to a side-terminal LLM (Gemini today, others next), return a dense markdown blueprint to Claude. Saves context window + token cost.
MIT -
llm-council-mcpgithub.com/NabilAttia123/llm-council-mcpupdated this year
Get a second opinion from multiple frontier LLMs at once. MCP server for Claude Code that queries GPT, Gemini, Claude, Grok, or any LLM hosted on OpenRouter simultaneously with peer review.
-
code-aidegithub.com/logicia32/code-aideupdated this year
Offload code reading / review / edits to a cheap LLM via MCP — returns concise results to keep your main agent's context lean.
MIT -
Multi-language code graph memory system for LLMs
npm Apache-2.0 -
Token-efficient code generation pipeline - parallel implementation with hosted LLM (Cerebras) for ~60% token savings. Includes MCP server.
MIT ★ 3 -
Code review by a council of four free LLMs on OpenRouter, as a CLI and an MCP server. Findings are ranked by how many reviewers raised them.
pypi NOASSERTION -
Local LLM does Claude's bulk work: reads logs and files, researches the web, writes test-gated code.
npm MIT ★ 2 -
🧠 MCP server for semantic code search, cross-chat memory & token-slim routing — cut LLM token usage 80-95%. 100% local, no API keys. Works with Claude Code, Cursor, Codex, Hermes.
MIT ★ 2 -
arborgithub.com/nikita-voronoy/arborupdated this year
Code navigation MCP server. Fits your entire codebase into an LLM context.
MIT ★ 7 -
cognis-enginegithub.com/buimanhtoan-it/cognisupdated this year
Software Cognition Engine — MCP-native context capsules for LLM coding agents.
pypi Apache Software License -
Long-term memory for AI agents with no LLM call: an MCP server and Claude Code plugin, plus local-first context compression for LLM apps.
npm MIT ★ 25 -
themecontextmcpgithub.com/bidfox/themecontextmcpupdated this year
ThemeContext MCP Server -- Up-to-date themes design token for LLMs and AI code editors
MIT ★ 2 -
sentinel-aigithub.com/MaxwellCalkin/sentinel-aiupdated this year
Real-time AI safety guardrails for LLM apps. 10 scanners: prompt injection, PII, harmful content, code vulnerabilities, obfuscation detection. Sub-ms latency. Python + TypeScript SDKs. MCP proxy. Claude Code hooks.
Apache-2.0 ★ 27 -
winnow-mcpupdated this year
MCP server for Winnow neural code pruning - reduces LLM context size 23-38%
npm MIT -
codenavgithub.com/oni-zerone/codenavupdated this year
Index a repository into a code graph and let LLMs explore it over MCP
cargo MIT -
mcp-server-code-reviewgithub.com/danielefavi/mcp-server-code-reviewupdated this year
MCP server for code reviews - connects LLMs to GitHub and GitLab to analyze Merge/Pull Requests and provide expert feedback
npm ISC -
@klitchevo/code-councilgithub.com/klitchevo/code-councilupdated this year
Multi-model AI code review server using OpenRouter - get diverse perspectives from multiple LLMs in parallel
npm MIT -
code-security-skillgithub.com/Chiehyii/code-security-skillupdated this year
Secure-coding knowledge base and MCP server for AI coding assistants, covering OWASP Top 10, ASVS, CWE, API, LLM, and supply-chain security.
-
dvad-agent-nativegithub.com/briankelley/dvad-agent-nativeupdated this year
Agent-native adversarial review — multi-LLM review as an MCP server + Claude Code skill.
pypi MIT -
Log Claude Code sessions to Opik, the open-source LLM observability and evaluation platform, built by Comet. Tracing, evaluation, and skills for observable AI applications.
go Apache-2.0 ★ 25 -
intent-codegithub.com/harsharahul/intent-codeupdated this year
An intent-aware code-knowledge index for LLM agents: index a repository once, query it by intent, and read only what's needed
pypi MIT -
Contract-driven coding workflow: the strong model plans and accepts, your local LLM does the coding. 25 MCP tools + 6 skills, 8 agent hosts (ZCode, Claude Code, Codex, VS Code, DSH, pi, omp, opencode). Local-first, no telemetry.
MIT