mcp-local-rag

Name: mcp-local-rag
Author: shinpr

11kinstalls/mo · npm351stars68forks

How to install mcp-local-rag?

shinpr/mcp-local-rag

$git clone https://github.com/shinpr/mcp-local-rag

Installs into the current project.

Summary

Local RAG for developers via MCP or CLI. Semantic search with keyword boost for exact technical terms — fully private, zero setup.

Files · 1

View on GitHub

README.md

1	<p align="center">
2	<img src="assets/banner.jpg" alt="MCP Local RAG — Search below the surface." width="600" />
3	</p>
4
5	# MCP Local RAG
6
7	[![GitHub stars](https://img.shields.io/github/stars/shinpr/mcp-local-rag?style=social)](https://github.com/shinpr/mcp-local-rag)
8	[![npm version](https://img.shields.io/npm/v/mcp-local-rag.svg)](https://www.npmjs.com/package/mcp-local-rag)
9	[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
10	[![TypeScript](https://img.shields.io/badge/TypeScript-6.0-blue.svg?logo=typescript&logoColor=white)](https://www.typescriptlang.org/)
11	[![MCP Registry](https://img.shields.io/badge/MCP-Registry-green.svg)](https://registry.modelcontextprotocol.io/)
12
13	Local RAG for developers via MCP or CLI.
14	Semantic search with keyword boost for exact technical terms — fully private, zero setup.
15
16	## Features
17
18	- Semantic search with keyword boost
19	Vector search first, then keyword matching boosts exact matches. Terms like `useEffect`, error codes, and class names rank higher—not just semantically guessed.
20
21	- Smart semantic chunking
22	Chunks documents by meaning, not character count. Uses embedding similarity to find natural topic boundaries—keeping related content together and splitting where topics change.
23
24	- Quality-first result filtering
25	Groups results by relevance gaps instead of arbitrary top-K cutoffs. Get fewer but more trustworthy chunks.
26
27	- Runs entirely locally
28	No API keys, no cloud, no data leaving your machine. Works fully offline after the first model download.
29
30	- Zero-friction setup
31	One `npx` command. No Docker, no Python, no servers to manage.
32	Use via MCP, CLI, or both. Optional [Agent Skills](#agent-skills) help AI assistants form better queries and interpret results.
33
34	## Quick Start
35
36	Set `BASE_DIR` to the folder you want to search (or `BASE_DIRS` for multiple roots — see [Configuration](#configuration)). Documents must live under one of the configured roots.
37
38	Add the MCP server to your AI coding tool:
39
40	For Cursor — Add to `~/.cursor/mcp.json`:
41	```json
42	{
43	"mcpServers": {
44	"local-rag": {
45	"command": "npx",
46	"args": ["-y", "mcp-local-rag"],
47	"env": {
48	"BASE_DIR": "/path/to/your/documents"
49	}
50	}
51	}
52	}
53	```
54
55	For Codex — Add to `~/.codex/config.toml`:
56	```toml
57	[mcp_servers.local-rag]
58	command = "npx"
59	args = ["-y", "mcp-local-rag"]
60
61	[mcp_servers.local-rag.env]
62	BASE_DIR = "/path/to/your/documents"
63	```
64
65	For Claude Code — Run this command:
66	```bash
67	claude mcp add local-rag --scope user --env BASE_DIR=/path/to/your/documents -- npx -y mcp-local-rag
68	```
69
70	Restart your tool, then start using it:
71
72	```
73	You: "Ingest api-spec.pdf"
74	Assistant: Successfully ingested api-spec.pdf (47 chunks created)
75
76	You: "What does the API documentation say about authentication?"
77	Assistant: Based on the documentation, authentication uses OAuth 2.0 with JWT tokens.
78	The flow is described in section 3.2...
79	```
80
81	Or use directly as CLI — no MCP server needed:
82
83	```bash
84	npx mcp-local-rag ingest ./docs/
85	npx mcp-local-rag query "authentication API"
86	```
87
88	That's it. No Docker, no Python, no server setup.
89
90	## Why This Exists
91
92	You want AI to search your documents—technical specs, research papers, internal docs. But most solutions send your files to external APIs.
93
94	Privacy. Your documents might contain sensitive data. This runs entirely locally.
95
96	Cost. External embedding APIs charge per use. This is free after the initial model download.
97
98	Offline. Works without internet after setup.
99
100	Code search. Pure semantic search misses exact terms like `useEffect` or `ERR_CONNECTION_REFUSED`. Keyword boost catches both meaning and exact matches.
101
102	Agent reality. In practice, many AI environments mainly use tool calling. CLI support and Agent Skills make the same workflows available even without full MCP integration.
103
104	## Usage
105
106	mcp-local-rag provides two interfaces: an MCP server for AI coding tools and a *CLI

People who liked this also viewed

6 picks

Type

ddgsA metasearch library that aggregates results from diverse web search servicesMCP ServersMay 20266.5M2.8k
firecrawl-mcp-server🔥 Official Firecrawl MCP Server - Adds powerful web scraping and search to Cursor, Claude and any other LLM clients.MCP ServersJul 2026480k7.0k
fliGoogle Flights MCP, CLI and Python LibraryMCP ServersJul 2026211k3.0k
sembleFast and Accurate Code Search for Agents. Uses ~98% fewer tokens than grep+readMCP ServersJul 2026140k5.7k
duckduckgo-mcp-serverA Model Context Protocol (MCP) server that provides web search capabilities through DuckDuckGo, with additional features for content fetching and parsing.MCP ServersJul 2026121k1.4k
wikipedia-mcpA Model Context Protocol (MCP) server that retrieves information from Wikipedia to provide context to LLMs.MCP ServersMay 2026113k274

mcp-local-rag

11kinstalls/mo · npm351stars68forks

View on GitHub

How to install mcp-local-rag?

shinpr/mcp-local-rag

$git clone https://github.com/shinpr/mcp-local-rag

Installs into the current project.

Summary

Local RAG for developers via MCP or CLI. Semantic search with keyword boost for exact technical terms — fully private, zero setup.

Files · 1

View on GitHub

README.md

1	<p align="center">
2	<img src="assets/banner.jpg" alt="MCP Local RAG — Search below the surface." width="600" />
3	</p>
4
5	# MCP Local RAG
6
7	[![GitHub stars](https://img.shields.io/github/stars/shinpr/mcp-local-rag?style=social)](https://github.com/shinpr/mcp-local-rag)
8	[![npm version](https://img.shields.io/npm/v/mcp-local-rag.svg)](https://www.npmjs.com/package/mcp-local-rag)
9	[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT)
10	[![TypeScript](https://img.shields.io/badge/TypeScript-6.0-blue.svg?logo=typescript&logoColor=white)](https://www.typescriptlang.org/)
11	[![MCP Registry](https://img.shields.io/badge/MCP-Registry-green.svg)](https://registry.modelcontextprotocol.io/)
12
13	Local RAG for developers via MCP or CLI.
14	Semantic search with keyword boost for exact technical terms — fully private, zero setup.
15
16	## Features
17
18	- Semantic search with keyword boost
19	Vector search first, then keyword matching boosts exact matches. Terms like `useEffect`, error codes, and class names rank higher—not just semantically guessed.
20
21	- Smart semantic chunking
22	Chunks documents by meaning, not character count. Uses embedding similarity to find natural topic boundaries—keeping related content together and splitting where topics change.
23
24	- Quality-first result filtering
25	Groups results by relevance gaps instead of arbitrary top-K cutoffs. Get fewer but more trustworthy chunks.
26
27	- Runs entirely locally
28	No API keys, no cloud, no data leaving your machine. Works fully offline after the first model download.
29
30	- Zero-friction setup
31	One `npx` command. No Docker, no Python, no servers to manage.
32	Use via MCP, CLI, or both. Optional [Agent Skills](#agent-skills) help AI assistants form better queries and interpret results.
33
34	## Quick Start
35
36	Set `BASE_DIR` to the folder you want to search (or `BASE_DIRS` for multiple roots — see [Configuration](#configuration)). Documents must live under one of the configured roots.
37
38	Add the MCP server to your AI coding tool:
39
40	For Cursor — Add to `~/.cursor/mcp.json`:
41	```json
42	{
43	"mcpServers": {
44	"local-rag": {
45	"command": "npx",
46	"args": ["-y", "mcp-local-rag"],
47	"env": {
48	"BASE_DIR": "/path/to/your/documents"
49	}
50	}
51	}
52	}
53	```
54
55	For Codex — Add to `~/.codex/config.toml`:
56	```toml
57	[mcp_servers.local-rag]
58	command = "npx"
59	args = ["-y", "mcp-local-rag"]
60
61	[mcp_servers.local-rag.env]
62	BASE_DIR = "/path/to/your/documents"
63	```
64
65	For Claude Code — Run this command:
66	```bash
67	claude mcp add local-rag --scope user --env BASE_DIR=/path/to/your/documents -- npx -y mcp-local-rag
68	```
69
70	Restart your tool, then start using it:
71
72	```
73	You: "Ingest api-spec.pdf"
74	Assistant: Successfully ingested api-spec.pdf (47 chunks created)
75
76	You: "What does the API documentation say about authentication?"
77	Assistant: Based on the documentation, authentication uses OAuth 2.0 with JWT tokens.
78	The flow is described in section 3.2...
79	```
80
81	Or use directly as CLI — no MCP server needed:
82
83	```bash
84	npx mcp-local-rag ingest ./docs/
85	npx mcp-local-rag query "authentication API"
86	```
87
88	That's it. No Docker, no Python, no server setup.
89
90	## Why This Exists
91
92	You want AI to search your documents—technical specs, research papers, internal docs. But most solutions send your files to external APIs.
93
94	Privacy. Your documents might contain sensitive data. This runs entirely locally.
95
96	Cost. External embedding APIs charge per use. This is free after the initial model download.
97
98	Offline. Works without internet after setup.
99
100	Code search. Pure semantic search misses exact terms like `useEffect` or `ERR_CONNECTION_REFUSED`. Keyword boost catches both meaning and exact matches.
101
102	Agent reality. In practice, many AI environments mainly use tool calling. CLI support and Agent Skills make the same workflows available even without full MCP integration.
103
104	## Usage
105
106	mcp-local-rag provides two interfaces: an MCP server for AI coding tools and a *CLI

People who liked this also viewed

6 picks

Type