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

<AgentInstructions>

## Submitting Feedback

If you encounter incorrect, outdated, or confusing documentation on this page, submit feedback:

POST https://docs.endorlabs.com/feedback

```json
{
  "path": "/introduction/docs-mcp-server/index",
  "feedback": "Description of the issue"
}
```

Only submit feedback when you have something specific and actionable to report.

</AgentInstructions>

# Documentation MCP server

> Connect the Endor Labs documentation MCP server to your AI tools for accurate, up-to-date answers about Endor Labs.

export const DocsMcpInstallButton = ({variant}) => {
  const MCP_SERVER_URL = 'https://docs.endorlabs.com/mcp';
  const MCP_SERVER_NAME = 'endor-docs';
  const MCP_SERVER_DISPLAY_NAME = 'Endor Labs Docs';
  const CURSOR_DEEPLINK_BASE = 'cursor://anysphere.cursor-deeplink/mcp/install';
  const CURSOR_PROTOCOL = 'cursor:';
  const VSCODE_DEEPLINK_BASE = 'vscode:mcp/install';
  const VSCODE_PROTOCOL = 'vscode:';
  const BUTTON_LABELS = {
    cursor: 'Install in Cursor',
    vscode: 'Install in VS Code'
  };
  const CURSOR_SVG = <svg viewBox="0 0 24 24" width="18" height="18" fill="currentColor" style={{
    marginRight: '0.5rem',
    flexShrink: 0
  }}>
      <path d="M11.503.131 1.891 5.678a.84.84 0 0 0-.42.726v11.188c0 .3.162.575.42.724l9.609 5.55a1 1 0 0 0 .998 0l9.61-5.55a.84.84 0 0 0 .42-.724V6.404a.84.84 0 0 0-.42-.726L12.497.131a1.01 1.01 0 0 0-.996 0M2.657 6.338h18.55c.263 0 .43.287.297.515L12.23 22.918c-.062.107-.229.064-.229-.06V12.335a.59.59 0 0 0-.295-.51l-9.11-5.257c-.109-.063-.064-.23.061-.23" />
    </svg>;
  const VSCODE_SVG = <svg viewBox="0 0 24 24" width="18" height="18" fill="currentColor" style={{
    marginRight: '0.5rem',
    flexShrink: 0
  }}>
      <path d="M17.583.063a1.5 1.5 0 0 0-1.032.392 1.5 1.5 0 0 0-.001 0L7.05 9.6 2.856 6.295a1 1 0 0 0-1.272.09l-1.2 1.143a1 1 0 0 0 0 1.44L4.2 12l-3.816 3.03a1 1 0 0 0 0 1.442l1.2 1.143a1 1 0 0 0 1.272.09L7.05 14.4l9.5 9.145a1.5 1.5 0 0 0 1.033.392c.205 0 .405-.042.59-.124l4.2-1.89a1.5 1.5 0 0 0 .877-1.364V2.44a1.5 1.5 0 0 0-.877-1.362l-4.2-1.891a1.5 1.5 0 0 0-.59-.124M18 7.39v9.22L10.97 12z" />
    </svg>;
  const VARIANT_ICONS = {
    cursor: CURSOR_SVG,
    vscode: VSCODE_SVG
  };
  const [isDark, setIsDark] = useState(false);
  const [error, setError] = useState('');
  useEffect(() => {
    const check = () => {
      const root = document.documentElement;
      setIsDark(root.dataset.theme === 'dark' || root.classList.contains('dark'));
    };
    check();
    const observer = new MutationObserver(check);
    observer.observe(document.documentElement, {
      attributes: true,
      attributeFilter: ['data-theme', 'class']
    });
    return () => observer.disconnect();
  }, []);
  const styles = {
    primaryButton: {
      padding: '0.75rem 1.5rem',
      background: 'linear-gradient(to bottom, rgba(50,225,140,0.92), rgba(30,190,110,0.92))',
      color: isDark ? '#000' : '#fff',
      border: '1.5px solid rgba(38,208,124,0.6)',
      borderRadius: '10px',
      cursor: 'pointer',
      fontSize: '0.9rem',
      display: 'inline-flex',
      alignItems: 'center',
      fontWeight: 500,
      textDecoration: 'none'
    },
    errorBox: {
      marginTop: '0.5rem',
      padding: '0.5rem 0.75rem',
      background: isDark ? '#2d0f0f' : '#fde8e8',
      border: '1px solid ' + (isDark ? '#5c1e1e' : '#f1c0c0'),
      borderRadius: '8px',
      color: isDark ? '#ff6659' : '#9a0007',
      fontSize: '0.8rem'
    }
  };
  const isValidProtocolUri = (uri, expectedProtocol) => {
    try {
      return new URL(uri).protocol === expectedProtocol;
    } catch {
      return false;
    }
  };
  const buildCursorDeeplinkUri = () => {
    const config = {
      url: MCP_SERVER_URL
    };
    const encodedConfig = btoa(JSON.stringify(config));
    return CURSOR_DEEPLINK_BASE + '?name=' + encodeURIComponent(MCP_SERVER_NAME) + '&config=' + encodeURIComponent(encodedConfig);
  };
  const buildVscodeDeeplinkUri = () => {
    const payload = {
      name: MCP_SERVER_DISPLAY_NAME + ' (' + MCP_SERVER_NAME + ')',
      type: 'http',
      url: MCP_SERVER_URL
    };
    return VSCODE_DEEPLINK_BASE + '?' + encodeURIComponent(JSON.stringify(payload));
  };
  const handleInstall = useCallback(() => {
    setError('');
    const isVscode = variant === 'vscode';
    const deeplinkUri = isVscode ? buildVscodeDeeplinkUri() : buildCursorDeeplinkUri();
    const expectedProtocol = isVscode ? VSCODE_PROTOCOL : CURSOR_PROTOCOL;
    if (!isValidProtocolUri(deeplinkUri, expectedProtocol)) {
      setError('Unable to generate a valid install link.');
      return;
    }
    globalThis.location.href = deeplinkUri;
  }, [variant]);
  return <div className="not-prose docs-mcp-install" style={{
    margin: '1rem 0',
    fontFamily: '-apple-system, BlinkMacSystemFont, "Segoe UI", Helvetica, Arial, sans-serif'
  }}>
      <button type="button" onClick={handleInstall} style={styles.primaryButton}>
        {VARIANT_ICONS[variant]}
        {BUTTON_LABELS[variant]}
      </button>
      {error && <div style={styles.errorBox}>{error}</div>}
    </div>;
};

The Endor Labs documentation MCP server gives your AI tools direct access to the complete Endor Labs documentation. Instead of relying on outdated training data, your AI assistant can search and read current documentation in real time.

Endor Labs hosts the documentation MCP server at `https://docs.endorlabs.com/mcp` using HTTP transport. You do not need authentication, API keys, or local installation.

<Note>
  This page covers the **documentation** MCP server for searching Endor Labs docs. For the **security scanning** MCP server that scans your code for vulnerabilities, see [MCP server](/secure-ai-coding/mcp-server).
</Note>

## What the documentation MCP server provides

The MCP server provides the following tools that AI applications use:

* **Search documentation**: Searches across all Endor Labs documentation to find relevant content, returning snippets with titles and direct links.
* **Query docs filesystem**: Reads and navigates a virtual filesystem of documentation pages using shell-style commands like `rg`, `cat`, `head`, and `tree`.

Your AI system decides when to leverage these tools and can independently query the documentation as it generates a response, even without a direct prompt to do so.

## Connect the documentation MCP server

<Tabs>
  <Tab title="Cursor">
    Click **Install in Cursor** to add the documentation MCP server.

    <DocsMcpInstallButton variant="cursor" />

    To manually configure the MCP server, add the following JSON to a `.cursor/mcp.json` file in the root of your repository.

    ```json theme={null}
    {
      "mcpServers": {
        "endor-docs": {
          "url": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```

    You can also use the command palette. Press `Cmd + Shift + P` (or `Ctrl + Shift + P` on Windows and Linux), search for **MCP: Add Server**, and enter the URL `https://docs.endorlabs.com/mcp`.
  </Tab>

  <Tab title="VS Code">
    Click **Install in VS Code** to add the documentation MCP server.

    <DocsMcpInstallButton variant="vscode" />

    To manually configure the MCP server, create a `.vscode/mcp.json` file in the root of your repository and add the following JSON.

    ```json theme={null}
    {
      "servers": {
        "endor-docs": {
          "type": "http",
          "url": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="Claude">
    Add the Endor Labs documentation MCP server as a custom connector in the Claude web interface.

    1. Go to **Settings** > **Connectors**.
    2. Select **Add custom connector**.
    3. Enter the following details:
       * **Name**: `Endor Labs Docs`
       * **URL**: `https://docs.endorlabs.com/mcp`
    4. Select **Add**.
    5. When using Claude, select the attachments button (the plus icon) and choose the **Endor Labs Docs** connector.

    For more information, refer to the [Claude MCP documentation](https://support.anthropic.com/en/articles/11175166-how-do-i-connect-mcp-servers-to-claude-ai).
  </Tab>

  <Tab title="Claude Code">
    Run the following command to add the documentation MCP server to Claude Code.

    ```bash theme={null}
    claude mcp add --transport http endor-docs https://docs.endorlabs.com/mcp
    ```

    Run the following command to verify the connection.

    ```bash theme={null}
    claude mcp list
    ```

    For more information, refer to the [Claude Code documentation](https://docs.anthropic.com/en/docs/agents-and-tools/claude-code/tutorials#set-up-model-context-protocol-mcp).
  </Tab>

  <Tab title="Gemini CLI">
    Run the following command to add the documentation MCP server to Gemini CLI.

    ```bash theme={null}
    gemini mcp add --transport http endor-docs https://docs.endorlabs.com/mcp
    ```

    Run the following command to verify the connection.

    ```bash theme={null}
    gemini mcp list
    ```

    For more information, refer to the [Gemini CLI documentation](https://github.com/google-gemini/gemini-cli).
  </Tab>

  <Tab title="OpenAI Codex">
    <Warning>
      The documentation MCP server does not require authentication. However, Codex automatically detects OAuth support on HTTP MCP servers and attempts an OAuth flow that fails. To work around this, provide a placeholder bearer token environment variable when adding the server.
    </Warning>

    Run the following commands to add the documentation MCP server to OpenAI Codex.

    ```bash theme={null}
    export ENDOR_DOCS_KEY="dummy"
    codex mcp add endor-docs --url https://docs.endorlabs.com/mcp --bearer-token-env-var ENDOR_DOCS_KEY
    ```

    Set the `ENDOR_DOCS_KEY` environment variable in your shell before starting Codex. Add `export ENDOR_DOCS_KEY="dummy"` to your shell profile (such as `~/.zshrc` or `~/.bashrc`) so it persists across sessions.

    Run the following command to verify the connection.

    ```bash theme={null}
    codex mcp list
    ```

    You can also use `/mcp` in the Codex TUI to view active MCP servers.

    Alternatively, add the server directly to `~/.codex/config.toml`:

    ```toml theme={null}
    [mcp_servers.endor-docs]
    url = "https://docs.endorlabs.com/mcp"
    bearer_token_env_var = "ENDOR_DOCS_KEY"
    enabled = true
    ```

    Ensure that you set the `ENDOR_DOCS_KEY` environment variable in your shell profile before starting Codex.
  </Tab>

  <Tab title="Devin">
    Add the Endor Labs documentation MCP server to Devin through the MCP Marketplace.

    1. Navigate to [Settings > MCP Marketplace](https://app.devin.ai/settings/mcp-marketplace) in Devin.
    2. Select **Add Your Own** to add a custom MCP server.
    3. Under **HTTP Configuration**, enter the URL `https://docs.endorlabs.com/mcp`.
    4. Select **Save Changes**.

    The documentation MCP server does not require secrets or authentication.
  </Tab>

  <Tab title="Augment Code">
    Add the Endor Labs documentation MCP server to Augment Code.

    1. Open the Augment Code extension in Visual Studio Code.
    2. Select the **Settings** icon in the upper right of the Augment panel.
    3. In the **MCP** section, select **Import from JSON** or **+** to add a server.
    4. Paste the following configuration:

    ```json theme={null}
    {
      "mcpServers": {
        "endor-docs": {
          "url": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```

    Alternatively, select **+**, set **Name** to `endor-docs`, and set the **URL** to `https://docs.endorlabs.com/mcp`.
  </Tab>

  <Tab title="IntelliJ IDEA">
    Add the Endor Labs documentation MCP server to IntelliJ IDEA through GitHub Copilot.

    1. Open **GitHub Copilot Chat** and switch to **Agent** mode.
    2. Select **Configure Tools**, then select **+ Add More Tools...** to open `mcp.json`.
    3. Add the following configuration:

    ```json theme={null}
    {
      "servers": {
        "endor-docs": {
          "type": "http",
          "url": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```

    4. Save and close `mcp.json`. Switch from **Agent** to **Ask** mode and back to **Agent** mode to reload the MCP server.
  </Tab>

  <Tab title="Antigravity">
    Add the Endor Labs documentation MCP server to Google Antigravity.

    1. On the agent panel, click **...**, then select **MCP Servers** > **Manage MCP Servers**.
    2. Click **View raw config** to open `mcp_config.json`.
    3. Add the following configuration:

    ```json theme={null}
    {
      "mcpServers": {
        "endor-docs": {
          "serverUrl": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```

    4. Save the file. Antigravity automatically reloads the MCP server configuration.

    For more information, refer to the [Google Antigravity documentation](https://antigravity.google/docs/mcp).
  </Tab>

  <Tab title="Other tools">
    For any MCP-compatible tool, use the following server URL.

    ```text theme={null}
    https://docs.endorlabs.com/mcp
    ```

    The documentation MCP server uses **HTTP transport**. Configure your tool with the URL above. You do not need authentication headers or API keys.

    If your tool uses a JSON configuration file, use one of the following formats.

    For tools that use `mcpServers` as the top-level key:

    ```json theme={null}
    {
      "mcpServers": {
        "endor-docs": {
          "url": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```

    For tools that use `servers` as the top-level key:

    ```json theme={null}
    {
      "servers": {
        "endor-docs": {
          "type": "http",
          "url": "https://docs.endorlabs.com/mcp"
        }
      }
    }
    ```
  </Tab>
</Tabs>

## Verify the connection

After connecting, test the MCP server by asking your AI tool a question about Endor Labs. For example:

* "How do I configure a scan profile in Endor Labs?"
* "What package managers does Endor Labs support?"
* "How do I set up GitHub Actions with Endor Labs?"

Your AI tool should search the Endor Labs documentation and provide an answer with references to specific documentation pages.

## Use with the security scanning MCP server

The documentation MCP server works alongside the [Endor Labs security scanning MCP server](/secure-ai-coding/mcp-server). You can connect both to the same AI tool.

* The **documentation MCP server** answers questions about Endor Labs features, configuration, and usage.
* The **security scanning MCP server** scans your code for vulnerabilities, leaked secrets, and security issues.

Your AI assistant selects the appropriate server based on the conversation context. Connecting both servers enables full access to Endor Labs capabilities.