Glean MCP Server: Scale & Secure Team Collaboration

Boost your Glean Chat performance with MCP Server – effortlessly scale, secure, and optimize for seamless collaboration. Power your team’s productivity!" )

Visit Repository

✨ Developer Tools

4.3(77 reviews)

115 saves

53 comments

71% of users reported increased productivity after just one week

About Glean MCP Server

What is Glean MCP Server: Scale & Secure Team Collaboration?

Imagine a superhero sidekick that bridges the gap between your team’s brainpower and Glean’s AI superpowers. The Glean MCP Server is your trusty middleware, connecting the Model Context Protocol (MCP) to Glean’s Chat API. It’s like having a digital middleman who ensures your team’s questions about 2025 holidays (or anything else) get answered efficiently—without leaving security vulnerabilities wide open.

How to Use Glean MCP Server: Scale & Secure Team Collaboration?

Ready to get your hands dirty? Let’s break it down:

1. Clone the repo and grab your dependencies like a pro: git clone then uv sync.

2. Configure your .env file like a secret agent—API keys and base URLs need to stay undercover.

3. Launch the server in standalone mode or debug like a wizard with the MCP inspector. And yes, testing with JSON payloads is encouraged (we won’t judge your holiday question obsession).

Glean MCP Server Features

Key Features of Glean MCP Server: Scale & Secure Team Collaboration?

Why’s this tool not just another cog in the wheel?

Scaling like a champ: Handles your team’s chaotic questions without melting down.
Security first: Locks down API access so only authorized folks get the goods.
Flexibility on tap: Works in standalone mode, debug mode, or even with Cursor—like a tech Swiss Army knife.

Use Cases of Glean MCP Server: Scale & Secure Team Collaboration?

Think of it as your team’s secret weapon for:

Real-time knowledge mining (bye-bye, endless email threads).
Automating repetitive queries so your team can focus on actual work.
Securely integrating with existing workflows without reinventing the wheel.

Glean MCP Server FAQ

FAQ from Glean MCP Server: Scale & Secure Team Collaboration?

Q: Why UV package manager?
UV keeps dependencies tidy, like a digital Marie Kondo. Plus, it’s built for MCP love.

Q: How do I debug without losing my sanity?
Use the MCP inspector—your new best friend for spotting errors faster than you can say “JSON payload.”

Q: What if my server acts up?
Check your Python version (3.10+ only!), verify the API key’s permissions, and consult the docs (we’ve all been there).

Need more? The MCP docs and Glean API guides are your next stops.

Content

Glean MCP Server

A Model Context Protocol (MCP) server that integrates with Glean's Chat API.

Prerequisites

Python 3.10+
UV package manager (recommended)
Glean API key with appropriate permissions(/rest/api/v1/*)

Installation

Clone this repository:

git clone https://github.com/rahul-roy-glean/glean-mcp-server.git

cd glean-mcp-server

Install dependencies using UV:

uv sync

uv lock

Configuration

Before running the server, you need to set up your Glean API credentials. Create a .env file in the project root with the following variables:

GLEAN_API_KEY=your_api_key_here
GLEAN_BASE_URL=https://your-domain-be.glean.com/rest/api/v1/

Running the Server

Standalone Mode

To run the server in standalone mode:

uv --directory <PATH_TO_CHECKOUT> run glean_server.py

Debug Mode

To debug the server with the MCP inspector:

npx @modelcontextprotocol/inspector uv --directory <PATH_TO_CHECKOUT> run glean_server.py

You can then test with JSON payloads like:

{
  "messages": [
    {
      "author": "USER",
      "fragments": [
        {
          "text": "What are the company holidays in 2025 ?"
        }
      ],
      "messageType": "CONTENT"
    }
  ],
  "saveChat": true,
  "stream": false
}

Integration with Cursor

To use this server with Cursor, add the following to ~/.cursor/mcp.json:

{
    "mcpServers": {       
        "glean": {
            "command": "uv",
            "args": [
                "--directory",
                "<PATH_TO_CHECKOUT>",
                "run", "glean_server.py"
            ]
        }
    }
}

Documentation Links