> ## Documentation Index
> Fetch the complete documentation index at: https://anaconda.com/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# Anaconda MCP

export const Comments = ({children}) => {
  return <div class="my-4 px-5 py-4 overflow-hidden rounded-2xl flex gap-3 border border-zinc-500/20 bg-zinc-50/50 dark:border-zinc-500/30 dark:bg-zinc-500/10" data-callout-type="comments">
      <div class="w-4">
        <svg width="14" height="14" viewBox="0 0 640 640" fill="currentColor" xmlns="http://www.w3.org/2000/svg" class="w-5 h-5" aria-label="Comments">
            <path d="M320 112C434.9 112 528 205.1 528 320C528 434.9 434.9 528 320 528C205.1 528 112 434.9 112 320C112 205.1 205.1 112 320 112zM320 576C461.4 576 576 461.4 576 320C576 178.6 461.4 64 320 64C178.6 64 64 178.6 64 320C64 461.4 178.6 576 320 576zM280 400C266.7 400 256 410.7 256 424C256 437.3 266.7 448 280 448L360 448C373.3 448 384 437.3 384 424C384 410.7 373.3 400 360 400L352 400L352 312C352 298.7 341.3 288 328 288L280 288C266.7 288 256 298.7 256 312C256 325.3 266.7 336 280 336L304 336L304 400L280 400zM320 256C337.7 256 352 241.7 352 224C352 206.3 337.7 192 320 192C302.3 192 288 206.3 288 224C288 241.7 302.3 256 320 256z" />
        </svg>
      </div>
      <div class="text-sm prose min-w-0 w-full">
        {children}
      </div>
    </div>;
};

export const GCell = ({children, className}) => <div className={`grid-table-cell ${className || ""}`} role="cell">
    {children}
  </div>;

export const GTH = ({children, className}) => <div className={`grid-table-th ${className || ""}`} role="columnheader">
    {children}
  </div>;

export const GRow = ({children}) => <div className="grid-table-row" role="row">{children}</div>;

export const GBody = ({children}) => <div className="grid-table-body" role="rowgroup">{children}</div>;

export const GHead = ({children}) => <div className="grid-table-head" role="rowgroup">{children}</div>;

export const GTable = ({children, className, cols}) => <div className={`grid-table not-prose overflow-hidden rounded-2xl ${className || ""}`} style={{
  "--grid-table-cols": cols
}} role="table">
    {children}
  </div>;

Anaconda MCP is a Model Context Protocol (MCP) server that gives AI coding assistants full awareness of your conda environments, packages, and channel configurations. Once connected, you can create environments, install packages, and manage dependencies using natural language directly from your AI coding assistant.

Anaconda MCP currently supports the following assistants: Claude Desktop, Claude Code, Cursor, VS Code, Opencode, and Windsurf.

## Prerequisites

You must have at least one supported AI coding assistant installed on your machine.

## Connecting Anaconda MCP

<Note>
  <Icon icon="plug" iconType="light" /> **MCP** is displayed in the left-hand navigation by default. Disable it on the [Settings](/docs/anaconda-desktop/settings#enabling-beta-features) page. If you belong to an organization, your administrator may restrict access to beta features.
</Note>

<Tabs>
  <Tab title="Anaconda Desktop">
    <Steps>
      <Step title="Configure your AI coding assistant">
        1. Select <Icon icon="plug" iconType="light" /> **MCP** under <Icon icon="flask" iconType="solid" /> **BETA** in the left-hand navigation.
        2. Toggle on the AI coding assistant you want to configure.

                   <Frame>
                     <img src="https://mintcdn.com/anaconda-29683c67/Y9GOZcy1bvac0fle/images/desktop-anaconda-mcp.png?fit=max&auto=format&n=Y9GOZcy1bvac0fle&q=85&s=a818c6aec4fec2c6acb138dcd9304a86" alt="Anaconda MCP setup page" width="3460" height="1792" data-path="images/desktop-anaconda-mcp.png" />
                   </Frame>

        Anaconda Desktop installs the `anaconda-mcp` package into a new [environment](/docs/anaconda-desktop/environments) called `anaconda-mcp` and configures the selected assistant automatically. A confirmation message appears when setup is complete.
      </Step>

      <Step title="Restart your AI coding assistant">
        Quit and restart your AI coding assistant for the changes to take effect.
      </Step>

      <Step title="Verify the connection">
        Ask your assistant something like "List my conda environments" to verify that Anaconda MCP is connected and responding.

        Alternatively, you can verify the MCP server is running by checking the settings in your AI coding assistant's configuration. For example, in Claude Desktop, navigate to **Settings** > **Developer** > **Local MCP Servers** and look for **anaconda-mcp**.

        <Tip>
          You can add "In conda" or "Using the conda server" to your prompts to ensure requests are routed to the Anaconda MCP server.
        </Tip>
      </Step>
    </Steps>
  </Tab>

  <Tab title="CLI">
    <Steps>
      <Step title="Open a shell application">
        Open Anaconda Prompt (Terminal on macOS/Linux).
      </Step>

      <Step title="Create a conda environment and install anaconda-mcp">
        ```sh theme={null}
        conda create --name <ENV_NAME> anaconda-mcp
        ```

        <Comments>
          Replace `<ENV_NAME>` with the name of the new environment.
        </Comments>

        <Tip>
          Anaconda recommends installing `anaconda-mcp` in its own dedicated environment rather than `base` to avoid Python version and dependency conflicts.
        </Tip>
      </Step>

      <Step title="Activate the environment">
        ```sh theme={null}
        conda activate <ENV_NAME>
        ```

        <Comments>
          Replace `<ENV_NAME>` with the name of the environment where you installed `anaconda-mcp`.
        </Comments>
      </Step>

      <Step title="Run the setup wizard">
        ```sh theme={null}
        anaconda mcp setup
        ```

        The wizard prompts you to select your AI coding assistant and whether to configure it at a global or project level. If you're not logged in to your anaconda.com account, the wizard prompts you to log in. You're also prompted to agree to the [Anaconda MCP Beta Agreement](https://www.anaconda.com/legal/terms/mcpbeta) the first time you run this command.
      </Step>

      <Step title="Restart your AI coding assistant">
        Quit and restart your AI coding assistant for the changes to take effect.
      </Step>
    </Steps>

    For full CLI reference documentation, see the [Anaconda MCP CLI reference](/docs/cli-reference/anaconda-mcp/getting-started).
  </Tab>
</Tabs>

## Capabilities

Anaconda MCP exposes the following conda environment and package management capabilities:

<GTable cols="40% 60%">
  <GHead>
    <GRow>
      <GTH>Capability</GTH>
      <GTH>Natural language prompt example</GTH>
    </GRow>
  </GHead>

  <GBody>
    <GRow>
      <GCell>**Create a new environment**</GCell>
      <GCell>"Create a new conda environment named `my-env` with Python 3.10 and numpy."</GCell>
    </GRow>

    <GRow>
      <GCell>**List all environments**</GCell>
      <GCell>"Show all my conda environments."</GCell>
    </GRow>

    <GRow>
      <GCell>**Show environment details**</GCell>
      <GCell>"What packages are installed in the `my-env` environment?"</GCell>
    </GRow>

    <GRow>
      <GCell>**Install, update, or remove packages**</GCell>
      <GCell>"Add pandas and matplotlib to `my-env`."</GCell>
    </GRow>

    <GRow>
      <GCell>**Search for packages**</GCell>
      <GCell>"Find all available versions of scikit-learn."</GCell>
    </GRow>

    <GRow>
      <GCell>**Remove an environment**</GCell>
      <GCell>"Delete the environment called `oldenv`."</GCell>
    </GRow>

    <GRow>
      <GCell>**Search packages on Anaconda.org**</GCell>
      <GCell>"Find me the latest version of `scikit-learn` available on the main channel."</GCell>
    </GRow>

    <GRow>
      <GCell>**Search notebooks and collections on nb.anaconda.com, pyscript.com, and Anaconda Toolbox**</GCell>
      <GCell>"Find my Jupyter notebook on nb.anaconda.com where I built a time series forecasting model so I can reuse that analysis."</GCell>
    </GRow>

    <GRow>
      <GCell>**Search Anaconda documentation**</GCell>
      <GCell>"Search the Anaconda docs to show me how to configure conda to use my organization channel as the only channel I install packages from."</GCell>
    </GRow>

    <GRow>
      <GCell>**Search the Anaconda community forum**</GCell>
      <GCell>"Search the Anaconda forum to see if anyone else has run into CUDA version conflicts when installing PyTorch."</GCell>
    </GRow>
  </GBody>
</GTable>