Atlan MCP Server

The Atlan Model Context Protocol server allows your AI agents to interact with Atlan services.

Quick Start

Generate Atlan API key by following the documentation.
Select one of the following approaches based on your preference:
- Install via Docker - Uses Docker containers (recommended)
- Install via uv - Uses UV package manager

[!NOTE] Make sure to replace <YOUR_API_KEY>, <YOUR_INSTANCE>, and <YOUR_AGENT_ID> with your actual Atlan API key, instance URL, and agent ID(optional) in the configuration file respectively.

Install via Docker

Prerequisites:

Follow the official Docker installation guide for your operating system
Verify Docker is running:
```
docker --version
```

Add to Claude Desktop

Go to Claude > Settings > Developer > Edit Config > claude_desktop_config.json and add:

{
  "mcpServers": {
    "atlan": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "ATLAN_API_KEY=<YOUR_API_KEY>",
        "-e",
        "ATLAN_BASE_URL=https://<YOUR_INSTANCE>.atlan.com",
        "-e",
        "ATLAN_AGENT_ID=<YOUR_AGENT_ID>",
        "ghcr.io/atlanhq/atlan-mcp-server:latest"
      ]
    }
  }
}

Add to Cursor

Open Cursor > Settings > Tools & Integrations > New MCP Server to include the following:

{
  "mcpServers": {
    "atlan": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "ATLAN_API_KEY=<YOUR_API_KEY>",
        "-e",
        "ATLAN_BASE_URL=https://<YOUR_INSTANCE>.atlan.com",
        "-e",
        "ATLAN_AGENT_ID=<YOUR_AGENT_ID>",
        "ghcr.io/atlanhq/atlan-mcp-server:latest"
      ]
    }
  }
}

Install via uv

Prerequisites:

Install uv:

# macOS/Linux
curl -LsSf https://astral.sh/uv/install.sh | sh

# Windows (PowerShell)
powershell -c "irm https://astral.sh/uv/install.ps1 | iex"

# Alternative: if you already have Python/pip
pip install uv

Verify installation:
```
uv --version
```

[!NOTE] With uv, uvx automatically fetches the latest version each time you run it. For more predictable behavior, consider using the Docker option.

Add to Claude Desktop

Go to Claude > Settings > Developer > Edit Config > claude_desktop_config.json to include the following:

{
  "mcpServers": {
    "atlan": {
      "command": "uvx",
      "args": ["atlan-mcp-server"],
      "env": {
        "ATLAN_API_KEY": "<YOUR_API_KEY>",
        "ATLAN_BASE_URL": "https://<YOUR_INSTANCE>.atlan.com",
        "ATLAN_AGENT_ID": "<YOUR_AGENT_ID>"
      }
    }
  }
}

Add to Cursor

Open Cursor > Settings > Tools & Integrations > New MCP Server to include the following:

{
  "mcpServers": {
    "atlan": {
      "command": "uvx",
      "args": ["atlan-mcp-server"],
      "env": {
        "ATLAN_API_KEY": "<YOUR_API_KEY>",
        "ATLAN_BASE_URL": "https://<YOUR_INSTANCE>.atlan.com",
        "ATLAN_AGENT_ID": "<YOUR_AGENT_ID>"
      }
    }
  }
}

Available Tools

Tool	Description
`search_assets`	Search for assets based on conditions
`get_assets_by_dsl`	Retrieve assets using a DSL query
`traverse_lineage`	Retrieve lineage for an asset
`update_assets`	Update asset attributes (user description and certificate status)
`create_glossaries`	Create glossaries
`create_glossary_categories`	Create glossary categories
`create_glossary_terms`	Create glossary terms

Tool Access Control

The Atlan MCP Server includes a configurable tool restriction middleware that allows you to control which tools are available to users. This is useful for implementing role-based access control or restricting certain operations in specific environments.

Restricting Tools

You can restrict access to specific tools using the RESTRICTED_TOOLS environment variable. Provide a comma-separated list of tool names that should be blocked:

Docker Configuration

{
  "mcpServers": {
    "atlan": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "ATLAN_API_KEY=<YOUR_API_KEY>",
        "-e",
        "ATLAN_BASE_URL=https://<YOUR_INSTANCE>.atlan.com",
        "-e",
        "ATLAN_AGENT_ID=<YOUR_AGENT_ID>",
        "-e",
        "RESTRICTED_TOOLS=get_assets_by_dsl_tool,update_assets_tool",
        "ghcr.io/atlanhq/atlan-mcp-server:latest"
      ]
    }
  }
}

uv Configuration

{
  "mcpServers": {
    "atlan": {
      "command": "uvx",
      "args": ["atlan-mcp-server"],
      "env": {
        "ATLAN_API_KEY": "<YOUR_API_KEY>",
        "ATLAN_BASE_URL": "https://<YOUR_INSTANCE>.atlan.com",
        "ATLAN_AGENT_ID": "<YOUR_AGENT_ID>",
        "RESTRICTED_TOOLS": "get_assets_by_dsl_tool,update_assets_tool"
      }
    }
  }
}

Available Tool Names for Restriction

You can restrict any of the following tools:

search_assets_tool - Asset search functionality
get_assets_by_dsl_tool - DSL query execution
traverse_lineage_tool - Lineage traversal
update_assets_tool - Asset updates (descriptions, certificates)
create_glossaries - Glossary creation
create_glossary_categories - Category creation
create_glossary_terms - Term creation

Common Use Cases

Read-Only Access

Restrict all write operations:

RESTRICTED_TOOLS=update_assets_tool,create_glossaries,create_glossary_categories,create_glossary_terms

Disable DSL Queries

For security or performance reasons:

RESTRICTED_TOOLS=get_assets_by_dsl_tool

Minimal Access

Allow only basic search:

RESTRICTED_TOOLS=get_assets_by_dsl_tool,update_assets_tool,traverse_lineage_tool,create_glossaries,create_glossary_categories,create_glossary_terms

How It Works

When tools are restricted:

Hidden from listings: Restricted tools won't appear when clients request available tools
Execution blocked: If someone tries to execute a restricted tool, they'll receive a clear error message
Logged: All access decisions are logged for monitoring and debugging

No Restrictions (Default)

If you don't set the RESTRICTED_TOOLS environment variable, all tools will be available by default.

Transport Modes

The Atlan MCP Server supports three transport modes, each optimized for different deployment scenarios. For more details about MCP transport modes, see the official MCP documentation.

Transport Mode	Use Case	Benefits	When to Use
stdio (Default)	Local development, IDE integrations	Simple, direct communication	Claude Desktop, Cursor IDE
SSE (Server-Sent Events)	Remote deployments, web browsers	Real-time streaming, web-compatible	Cloud deployments, web clients
streamable-http	HTTP-based remote connections	Standard HTTP, load balancer friendly	Kubernetes, containerized deployments

For comprehensive deployment instructions, configuration examples, and production best practices, see our Deployment Guide.

Production Deployment

Host the Atlan MCP container image on the cloud/platform of your choice
Make sure you add all the required environment variables
Choose the appropriate transport mode for your deployment scenario. SSE Transport is recommended for production (-e MCP_TRANSPORT=sse)
For detailed deployment scenarios and configurations, refer to the Deployment Guide

Remote MCP Configuration

We currently do not have a remote MCP server for Atlan generally available.

You can use the mcp-remote local proxy tool to connect it to your remote MCP server.

This lets you to test what an interaction with your remote MCP server will be like with a real-world MCP client.

{
  "mcpServers": {
    "math": {
      "command": "npx",
      "args": ["mcp-remote", "https://hosted-domain"]
    }
  }
}

Develop Locally

Want to develop locally? Check out our Local Build Guide for a step-by-step walkthrough!

Need Help?

Reach out to [email protected] for any questions or feedback
You can also directly create a GitHub issue and we will answer it for you

Frequently Asked Questions

Do I need Python installed?

Short answer: It depends on your installation method.

Docker (Recommended): No Python installation required on your host machine. The container includes everything needed.
uv: A Python runtime is needed, but uv will automatically download and manage Python 3.11+ for you if it's not already available.

Technical details: The Atlan MCP server is implemented as a Python application. The Model Context Protocol itself is language-agnostic, but our current implementation requires Python 3.11+ to run.

Troubleshooting

If Claude Desktop shows an error similar to spawn uv ENOENT {"context":"connection","stack":"Error: spawn uv ENOENT\n at ChildProcess._handle.onexit, it is most likely this issue where Claude is unable to find uv. To fix it:
- Make sure uv is installed and available in your PATH
- Run which uv to verify the installation path
- Update Claude's configuration to point to the exact uv path by running whereis uv and use that path

Atlan

Atlan MCP Server

Quick Start

Install via Docker

Add to Claude Desktop

Add to Cursor

Install via uv

Add to Claude Desktop

Add to Cursor

Available Tools

Tool Access Control

Restricting Tools

Docker Configuration

uv Configuration

Available Tool Names for Restriction

Common Use Cases

Read-Only Access

Disable DSL Queries

Minimal Access

How It Works

No Restrictions (Default)

Transport Modes

Production Deployment

Remote MCP Configuration

Develop Locally

Need Help?

Frequently Asked Questions

Do I need Python installed?

Troubleshooting

Be the First to Experience MCP Now