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.
-
mcp-local-raggithub.com/milliman-cmhh/mcp-local-ragupdated this year
Local MCP server for RAG over PDFs, DOCX, images, and plaintext files.
pypi AGPL-3.0-or-later -
Local-first multi-source document indexing and MCP retrieval for trusted agents
NOASSERTION -
local-raggithub.com/Ricardo-Kaminski/local-ragupdated this year
100% local RAG for Obsidian, Zotero, and Claude Code — LightRAG + Ollama + MCP
pypi MIT ★ 2 -
This is my attempt at creating a local RAG mcp server to chat with local docx, xlsx files.
★ 2 -
local-rag-coregithub.com/linkwut-create/local-rag-coreupdated this year
Headless local-first RAG knowledge library for MCP, CLI, HTTP, and AI-tool workflows.
MIT -
@iflow-mcp/patakuti-local-knowledge-rag-mcpupdated this year
Local Knowledge RAG MCP Server - Semantic search and retrieval for local documents using vector embeddings
npm MIT -
This is a local rag with retrieval-augmented experience log for coding with indexed framework, reranking documents, emdedding-normalization
MIT -
mcp-local-rag-anythinggithub.com/lzfxxx/mcp-local-rag-anythingupdated this year
Local RAG MCP Server with extended file support (including PPTX)
npm MIT -
iflow-mcp_mcp-local-raggithub.com/nkapila6/mcp-local-ragupdated this year
MCP "server" that can perform a web search locally without the use of APIs.
pypi MIT -
local-rag-mcpgithub.com/overlorde/local-rag-mcpupdated this year
MCP server with local RAG pipeline — give AI agents persistent codebase search using Ollama + ChromaDB. Zero cost, zero data leakage.
★ 1 -
@13w/local-raggithub.com/13w/local-ragupdated this year
Distributed semantic memory + code RAG as an MCP plugin for Claude Code agents
npm -
Local RAG MCP Server - Easy-to-setup document search with minimal configuration
npm MIT -
Local RAG MCP Server for opencode - with automatic MCP instructions for project/common indexing
npm MIT -
Local-first GraphRAG over a markdown knowledge base: LightRAG, Ollama/Qwen, MCP server, LLM router, scheduled self-maintenance
MIT -
@yikizi/mcp-local-raggithub.com/yikizi/mcp-local-ragupdated this year
Local RAG MCP Server - Enhanced with hybrid search, memory management, and code file support (fork of shinpr/mcp-local-rag)
npm MIT -
Local-Doc-Insight-RAG-MCP-Server-github.com/maryna2005/Local-Doc-Insight-RAG-MCP-Server-updated this year
A high-performance local RAG server built on Model Context Protocol (MCP), using FAISS and Sentence-Transformers for secure document analysis.
-
Local-first RAG server for developers. Semantic + keyword search for code and technical docs. Works with MCP or CLI. Fully private, zero setup.
npm MIT ★ 432 -
rag-localgithub.com/MauricioPerera/rag-localupdated this year
Local-first RAG memory for LLMs: Node server with ONNX embeddings, OKF-structured chunks, portable vector bundles, and REST + Web UI + CLI + MCP interfaces
MIT -
local-rag-mcpgithub.com/thewinci/local-rag-mcpupdated this year
Semantic search for your codebase — local-first RAG MCP server with hybrid search, AST-aware chunking, and usage analytics
npm MIT -
@winci/local-raggithub.com/thewinci/local-ragupdated this year
Semantic search for your codebase — local-first RAG MCP server with hybrid search, AST-aware chunking, and usage analytics
npm Apache-2.0 -
Local RAG MCP Server - Easy-to-setup document search with minimal configuration (fork with remote embedding server support)
npm MIT -
Read-only MCP server for semantic search over a local document corpus — on-device embeddings (Ollama), local Chroma store, symlink-hardened file access. Sibling of claude-code-session-memory.
MIT -
A semantic search and retrieval system for local documents using vector embeddings. Powered by MCP (Model Context Protocol).
MIT ★ 9 -
@damoqiongqiu/mcp-local-raggithub.com/damoqiongqiu/mcp-local-ragupdated this year
Code intelligence for AI coding assistants — AST-level semantic code search with keyword boost. Index your codebase, search by meaning, keep everything local.
npm MIT -
nexus-local-rag-mcpgithub.com/halilibrahimyesirci/nexus-local-rag-mcpupdated this year
Offline MCP server for local document search — semantic and keyword search over your PDFs, Markdown and text, with no API keys, no cloud or Docker.
MIT ★ 1 -
local-rag-mcpgithub.com/radif-ru/local-rag-mcpupdated this year
Локальная privacy-first база знаний на RAG и MCP: гибридный поиск (BM25 + векторный, слияние RRF), cross-encoder reranker, расширение запросов, FAISS, ответы локальной LLM через Ollama. Инструменты MCP (read / list / search) модель вызывает сама. Без облачных API.
-
local-MCP-RAGgithub.com/manali2304/local-MCP-RAGupdated this year
The MCP tool can then be used inside VS Code / MCP-enabled clients to search local internal documentation using natural language. Uses persistent embedding using minilm-l6-v2, supports multi format embedding and semantic retrieval .
-
Production-oriented RAG and Agentic AI platform with hybrid retrieval, GraphRAG, Knowledge Graphs, ontologies, MCP, LangGraph, multi-agent workflows, context engineering, secure multi-tenancy, RAGAS/MLflow evaluation, OpenTelemetry, Redis, Kubernetes, and Helm.
NOASSERTION -
Local semantic + keyword search across your Obsidian vault and source repos. LanceDB + Ollama (bge-m3), exposed to Claude Code and Cowork over MCP. Privacy-first: no cloud, no API keys, no telemetry.
MIT -
local-RAG-backendgithub.com/suwa-sh/local-RAG-backendupdated this year
This is the backend for a RAG system that runs on Docker Compose. It registers documents in a wide range of file formats, which can be searched using the MCP server.
MIT ★ 1 -
Local-first hybrid-RAG MCP server (Qdrant + BM25 via RRF), multi-step agent, offline Ollama fallback
MIT -
pdf-rag-mcpgithub.com/AstralApart/pdf-rag-mcpupdated this year
Local PDF RAG MCP server with SiliconFlow embeddings.
MIT -
Local hybrid RAG service for Obsidian vaults - MCP + REST, fail-closed secrets gate
-
localbrain-raggithub.com/sinwoo0225/localbrainupdated this year
Local-first general-purpose RAG with an MCP server, CLI, and web console
pypi MIT License -
Local-first document and media RAG with pgvector, Ollama, MCP, and optional OpenClaw integration
Apache-2.0 -
mcp-rag-servergithub.com/agarwalvishal/mcp-rag-serverupdated this year
MCP RAG server — local embeddings, your docs never leave your machine. Private knowledge base + web search for Claude, Cursor, and Ollama. Drop your docs, connect your AI client, done.
MIT ★ 6 -
Local-first RAG over engineering documents (PDF/DOCX/DOC/MD/TXT), exposed via MCP to Cursor/Claude and so on. MIT.
MIT ★ 1 -
rag-vaultgithub.com/maxrotemberg04-spec/rag-vaultupdated this year
Local-first RAG over my Obsidian vault — hand-rolled hybrid retrieval (BM25 + embeddings + RRF), cited answers, MCP server for AI agents.
MIT -
rag-bge-tokenisergithub.com/bengtfrost/rag-bge-tokeniserupdated this year
High-performance Local RAG server using MCP, BGE-M3, and SQLite-vec. Optimized for legal documents and exact token chunking.
NOASSERTION -
MCP server that gives a local LLM agentic RAG tools: hybrid search in Qdrant, read-only SQL over tables, calculator and time; Docker Compose
-
Local, offline knowledge base for papers and source code: hybrid RAG (sqlite-vec + BM25 + reranker) served over MCP to LM Studio, Continue and opencode; macOS and Linux
MIT -
Minimal local RAG MCP server: semantic search over your markdown docs, embeddings via Ollama, nothing leaves your machine.
npm MIT -
content-raggithub.com/Awis13/content-ragupdated this year
Local-first RAG service with glass-box hybrid retrieval (pgvector + Postgres FTS + RRF), cited answers, RAGAS eval, and an MCP server. Flask + Ollama, fully Dockerised.
MIT -
hybrid-rag-mcpgithub.com/LynncX/hybrid-rag-mcpupdated this year
a rag mcp supporting for both ways of local model and api . base on https://github.com/shinpr/mcp-local-rag.
MIT ★ 1 -
Local-first, offline-capable Agent memory & knowledge service. REST + MCP dual-protocol, hybrid retrieval (vector + BM25/RRF), document/webpage RAG. No LLM required for storage & retrieval
pypi Apache-2.0 ★ 5 -
A high-performance, 100% local Retrieval-Augmented Generation (RAG) system built on the Model Context Protocol (MCP). Designed specifically to help AI agents process complex scientific literature with unprecedented accuracy.
MIT ★ 1 -
PlayWright-RAG-CustomMCPgithub.com/pruthive580/PlayWright-RAG-CustomMCPupdated this year
Take a Jira ticket to a passing Playwright web-UI test, fully local: checks existing coverage, then runs it (env-aware) or authors a framework-standard test and self-heals until green. Custom RAG MCP + Playwright MCP + a token-slimming adapter. Driver-agnostic (local LLM / Claude Code / any model), 24GB laptop, no cloud.
★ 1 -
Local-first RAG engine with hybrid search, BGE reranker, self-learning loop, and MCP server
-
This is a local LLM provider, include rag-mcp and agent features, can load HF/gguf/onnx model formats..
Apache-2.0 -
Local RAG server for Claude Code, Cursor and other MCP clients: hybrid BM25 + vector search with cross-encoder reranking on by default, 13 MCP tools, 33 file types including code, re-indexing on file changes. One Python process with embedded ChromaDB and FastEmbed ONNX: no Docker, database server or API key; offline after the first model download.
npm MIT ★ 306 -
local-memory-mcpgithub.com/ptobey/local-memory-mcpupdated this year
Persistent local memory for Claude and ChatGPT. No cloud, no subscription. Self-hosted RAG over MCP.
MIT ★ 10 -
second-braingithub.com/tomnguyen103/second-brainupdated this year
Local-first personal AI workspace for cited RAG, Agentic RAG, hybrid pgvector search, source management, briefings, and MCP actions.
-
High-performance RAG server in Rust. Hybrid BM25 + vector search, dual rerankers (BGE/Mxbai), exact BGE-M3 tokenization, MCP server for Goose/Aider, LRU query cache, 100% local. Multi-arch container. 800% iGPU speedup on Void Linux + Intel Vulkan.
★ 4 -
Servidor MCP + agente de IA 100% local (Qwen via Ollama) com SQL seguro, RAG avaliado e evals automatizadas
MIT -
Local-first RAG knowledge base: hybrid search (FTS5 + vector), AI Q&A, knowledge graph, read-only MCP. Built with Next.js + SQLite.
MIT -
Local-first RAG engine built on the Laya decision engine: hybrid BM25 + dense recall, calibrated relevance re-ranking and answerability gating that return evidence with probabilities rather than generated prose. CLI, FastAPI HTTP API, MCP server and web GUI.
Apache-2.0 -
A lightweight agentic RAG-style chatbot and Model Context Protocol (MCP) server that answers questions about LangChain using local vector retrieval, real-time `llms.txt` document resolution, and Corrective RAG (CRAG) context quality evaluation. Built with LangChain, Chroma, Ollama, FastMCP, and Streamlit.
MIT -
vibe-megithub.com/borisfaj/vibe_meupdated this year
Local RAG system with MCP server
pypi MIT -
UI App for ChromaDB to build local RAG pipelines. Built for macOS, native Swift. Gives you full control over embedding for RAG.
MPL-2.0 ★ 2 -
pow-rag-mcpupdated this year
Local RAG MCP server for code and documentation
pypi Apache-2.0