Verified

Graphlit MCP Server

Ingest and search knowledge across tools like Slack, Jira, and GitHub for AI RAG workflows.

Data last scanned today · Reviewed today

12 4 dangerous tools67 tools★ 379TypeScriptMIT

Overview

The Graphlit MCP Server facilitates a RAG-ready knowledge base by connecting external data sources to your MCP-compatible AI clients. It supports ingestion from a wide array of platforms including Notion, Linear, Google Drive, and Discord, processing files and web pages into searchable markdown and transcribing audio or video content. Beyond data ingestion, the server includes built-in tools for web crawling, search, and LLM-based conversation. Users can also perform operational tasks such as managing project collections, extracting structured JSON, and triggering notifications across various channels like Slack or email.

Our verdict

This server carries a high risk level with a security score of 37/100. It is a community-maintained project rather than an official Graphlit release, and its authentication methods and read-only mode safety have not been editorially reviewed. Due to these factors, it is best suited for experimental environments rather than sensitive production workflows.

Supports a broad range of data ingestion and publication tools.
Built-in web crawling and search capabilities eliminate external tool dependencies.
Includes only two direct dependencies for a reduced supply-chain surface.
Open source and licensed under the MIT license.

High risk score due to lack of official maintenance.
Security-critical features like authentication have not undergone external review.
Repository has not received updates for over 200 days.

Setup

Claude Desktop

{
  "mcpServers": {
    "graphlit-mcp-server": {
      "args": [
        "-y",
        "graphlit-mcp-server"
      ],
      "command": "npx"
    }
  }
}

Paste into ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows), then restart Claude Desktop. Some servers require API keys or tokens as environment variables — check the project's README and add an "env" object if needed.

No read-only mode — running this grants full read/write access.
No built-in authentication — anyone with access to this config can invoke every tool it exposes.

Tools

Tool	Description	Risk
Query Contents	Query contents.	low
Query Collections	Query collections.	low
Query Feeds	Query feeds.	low
Query Conversations	Query conversations.	low
Retrieve Relevant Sources	Retrieve relevant sources.	low
Retrieve Similar Images	Retrieve similar images.	low
Visually Describe Image	Visually describe image.	low
Prompt LLM Conversation	Prompt LLM conversation.	low
Extract Structured JSON from Text	Extract structured JSON from text.	low
Publish as Audio (ElevenLabs Audio)	Publish as audio.	medium