DevRag - Lightweight Local RAG MCP Server

Created By
tomohiro-owada6 months ago
Lightweight local RAG MCP server for semantic vector search over markdown documents. Reduces token consumption by 40x. Supports Japanese and English.
Overview

what is DevRag?

DevRag is a lightweight local RAG MCP server designed for semantic vector search over markdown documents, significantly reducing token consumption and improving search efficiency.

how to use DevRag?

To use DevRag, run the server locally and utilize its tools to index markdown files, perform semantic searches, and manage indexed documents without incurring API costs.

key features of DevRag?

  • 40x token reduction via semantic vector search
  • 15x faster responses with ~95ms search latency
  • Fully local operation with no API costs
  • Multi-language support for Japanese and English
  • Filtered search capabilities by directory and filename patterns
  • Single binary with no dependencies

use cases of DevRag?

  1. Efficiently searching through large collections of markdown documents.
  2. Reducing operational costs for applications requiring semantic search.
  3. Supporting multi-language documentation management.

FAQ from DevRag?

  • Can DevRag handle multiple languages?

Yes! DevRag supports both Japanese and English.

  • Is there any cost associated with using DevRag?

No, DevRag runs entirely locally, eliminating API costs.

  • How fast are the search responses?

DevRag provides responses in approximately 95ms, making it very efficient.

Server Config

{
  "mcpServers": {
    "devrag": {
      "type": "stdio",
      "command": "/usr/local/bin/devrag"
    }
  }
}
Project Info
Created At
6 months ago
Updated At
6 months ago
Author Name
tomohiro-owada
Star
-
Language
-
License
-

Recommend Servers

View All
Crevio

a day ago
Bring your real authenticated browser session to AI coding agents. Local-first MCP server + Chrome MV3 extension. No cloud. No telemetry.
@Cubenest

peek records the user's actual logged-in browser (DOM via rrweb, console events, network metadata, optional response bodies via opt-in Deep capture) through a Chrome MV3 extension. The extension ships events through a native-messaging stdio bridge to a local MCP server (peek-mcp), which persists them to a SQLite database at ~/.peek/sessions.db. AI coding agents (Claude Code, Cursor, Cline, Windsurf) read sessions from the database via 10 MCP tools: Tool What it does list_recent_sessions List recently recorded sessions (id, origin, ts, event count). get_session_summary LLM-readable narrative summary of a session. get_session_console_errors Console errors recorded in a session. get_session_network_errors Failed/notable network requests in a session. get_user_action_before_error Last N user actions before a console error. generate_playwright_repro Generate a runnable Playwright test from a session. get_dom_snapshot Reconstruct the DOM at a given timestamp. query_dom_history Timeline of attribute/text changes for a selector. request_authorization Side-panel consent for write actions (Level 3). execute_action Dispatch a UI action (gated by permission level + destructive blocklist). Why local-first matters Every other "browser session for AI" tool ships to a vendor cloud. peek's SQLite + extension live on the user's machine — no remote endpoints, no telemetry. The privacy policy (docs/peek/PRIVACY_POLICY.md) is the source of truth. Install # 1. Add the MCP server to Claude Code claude mcp add peek -- npx -y @peekdev/mcp # 2. Install the Chrome extension from the Chrome Web Store # (link added once the CWS listing is approved)

19 hours ago