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

# Get started with Algolia Productivity MCP

> Connect LLM tools to your Algolia account with user-scoped access for internal workflows.

export const InstallButtonLabel = ({label}) => <span className="inline-flex items-center gap-2">
    <Icon icon="square-terminal" color="#ffffff" />
    <span>{label}</span>
  </span>;

export const installButtonClassName = "inline-flex items-center gap-2 whitespace-nowrap font-medium text-white bg-primary-dark hover:opacity-90 px-[14px] py-2 text-sm no-underline";

export const installButtonStyle = {
  borderRadius: "8px",
  borderBottom: "none"
};

export const getVSCodeInstallHref = (name, url) => `vscode:mcp/install?${encodeURIComponent(JSON.stringify({
  name,
  type: "http",
  url
}))}`;

export const getCursorInstallHref = (name, url) => `cursor://anysphere.cursor-deeplink/mcp/install?name=${name}&config=${btoa(JSON.stringify({
  url
}))}`;

Use Algolia Productivity MCP for exploration and analysis by people in your organization across the Algolia applications and indices allowed by their permissions.
To expose a curated set of indices to external or customer-facing AI assistants instead, use [Algolia Public MCP](/doc/guides/model-context-protocol/public-mcp).
To compare both, see the [MCP overview](/doc/guides/model-context-protocol).

Algolia Productivity MCP provides read-only access.
It can search and analyze your data, but it can't create, update, or delete indices, records, or settings.

## Before you begin

To use Algolia Productivity MCP, make sure you have:

* An Algolia account with access to the applications and indices you want to work with.
* An MCP client such as ChatGPT, Claude, Claude Code, Cursor, Gemini CLI, VS Code, or OpenAI Playground.

## Enable Algolia Productivity MCP

Productivity MCP is enabled once for the account.
Each user then signs in separately, and their Algolia permissions determine what they can access.

1. Go to Algolia dashboard > Generative AI > MCP Servers > Productivity
2. Enable Algolia Productivity MCP for your account

<Note>
  Keep the Productivity MCP enabled while you use MCP clients.
  Disabling it stops connected workflows immediately.
</Note>

## Connect from your MCP client

Open your MCP client and add a new MCP server or connector. Use the Productivity MCP URL
from the Algolia dashboard, then sign in when prompted so the MCP can inherit your Algolia permissions.

<Note>
  You don't need to provide an OAuth client ID or client secret.
  If your MCP client asks for these fields, leave them empty.
</Note>

<Tabs>
  <Tab title="ChatGPT">
    1. Open [Connectors settings](https://chatgpt.com/#settings/Connectors), then click **Create**.
    2. Fill in the following:
       * **Server URL:** `https://mcp.algolia.com/mcp`
       * **Label:** Algolia
       * **Description:** add a description that helps the LLM understand the tool's purpose.
       * **Authentication:** specify OAuth
    3. Check **I trust this application**.
    4. Click **Create** and sign in when prompted.
    5. Enable the Algolia MCP connection from the prompt bar.
  </Tab>

  <Tab title="Claude">
    1. Open [Connectors settings](https://claude.ai/new#settings/customize-connectors), click **+**, then click **Add custom connector**.
    2. Fill in the following:
       * **Remote MCP server URL:** `https://mcp.algolia.com/mcp`
       * **Name:** Algolia
    3. Click **Add** and sign in when prompted.
    4. Enable the Algolia MCP connection from the prompt bar.

    On Claude Team and Enterprise plans, an organization owner adds the connector from **Organization settings > Connectors**.
    Other members then click **Connect** on it in **Customize > Connectors**.
  </Tab>

  <Tab title="Claude Code">
    1. Run the following command in your terminal to register the MCP server:
       ```sh icon=square-terminal theme={"system"}
       claude mcp add --transport http algolia https://mcp.algolia.com/mcp
       ```
    2. Start a Claude Code session and run `/mcp` to authenticate with Algolia.
    3. Sign in when prompted to authorize Algolia.
  </Tab>

  <Tab title="Codex">
    Configure the Algolia MCP server with the CLI or in `~/.codex/config.toml`, then run `codex mcp login algolia` and sign in when prompted to authorize Algolia.

    * Run the following command in your terminal:
      ```sh icon=square-terminal theme={"system"}
      codex mcp add algolia --url https://mcp.algolia.com/mcp
      ```
    * Or add the following to your `~/.codex/config.toml` file:
      ```toml ~/.codex/config.toml icon=cog theme={"system"}
      [mcp_servers.algolia] # [!code ++:2]
      url = "https://mcp.algolia.com/mcp"
      ```
  </Tab>

  <Tab title="Cursor">
    <a href={getCursorInstallHref("algolia", "https://mcp.algolia.com/mcp")} target="_self" style={installButtonStyle} className={installButtonClassName}>
      <InstallButtonLabel label="Add to Cursor" />
    </a>

    To add the MCP server manually instead:

    1. Open **Settings** > **Cursor Settings**, then click **Tools & MCPs**.
    2. Click **Add Custom MCP**.
    3. Paste the MCP server configuration into the `mcp.json` file that opens, then save your changes.
       ```json .cursor/mcp.json icon=braces theme={"system"}
       {
         "mcpServers": {
           "algolia": { // [!code ++:3]
             "url": "https://mcp.algolia.com/mcp"
           }
         }
       }
       ```
    4. Click **Connect**, then sign in when you're prompted.

    Alternatively, you can add the same configuration to `~/.cursor/mcp.json` on macOS and Linux, or to `%USERPROFILE%\.cursor\mcp.json` on Windows.
  </Tab>

  <Tab title="VS Code">
    <a href={getVSCodeInstallHref("algolia", "https://mcp.algolia.com/mcp")} target="_self" style={installButtonStyle} className={installButtonClassName}>
      <InstallButtonLabel label="Add to VS Code" />
    </a>

    To add the MCP server manually instead:

    1. Open the Command Palette (<kbd>⇧⌘P</kbd> on macOS, <kbd>Ctrl+Shift+P</kbd> on Windows and Linux) and run **MCP: Add Server**.
    2. Select **HTTP** as the transport.
    3. Enter the **Server URL**: `https://mcp.algolia.com/mcp`
    4. Set the **Server ID** to `algolia` (or your preferred name).
    5. Choose **Workspace** or **Global** as the target.

    Alternatively, add the following to your `~/.vscode/mcp.json` file on macOS and Linux,
    or to `%USERPROFILE%\.vscode\mcp.json` on Windows:

    ```json ~/.vscode/mcp.json icon=braces theme={"system"}
    {
      "servers": {
        "algolia": { // [!code ++:4]
          "type": "http",
          "url": "https://mcp.algolia.com/mcp"
        }
      }
    }
    ```
  </Tab>

  <Tab title="Gemini CLI">
    Configure the Algolia MCP server with the CLI or in `~/.gemini/settings.json`, then sign in when prompted to authorize Algolia.

    * Run the following command in your terminal:
      ```sh icon=square-terminal theme={"system"}
      gemini mcp add algolia https://mcp.algolia.com/mcp -s user -t http
      ```
    * Or add the following to your `~/.gemini/settings.json` file:
      ```json ~/.gemini/settings.json icon=braces theme={"system"}
      {
        "mcpServers": {
          "algolia": { // [!code ++:3]
            "httpUrl": "https://mcp.algolia.com/mcp"
          }
        }
      }
      ```
  </Tab>
</Tabs>

## What you can do

Algolia Productivity MCP is designed for internal, user-scoped workflows:

* Run searches across the applications and indices you can access.
* Ask business intelligence-style questions like top searches, no-results queries, and trends.
* Explore index data and validate changes before making updates elsewhere.

## Tools exposed by Algolia Productivity MCP

Algolia Productivity MCP exposes a **tool** per index and analytics-specific tools.

| Tool                                            | API                                                                                   |
| ----------------------------------------------- | ------------------------------------------------------------------------------------- |
| **Search**                                      |                                                                                       |
| `algolia_search_list_indices`                   | [List indices](/doc/rest-api/search/list-indices)                                     |
| `algolia_search_index`                          | [Search an index](/doc/rest-api/search/search-single-index)                           |
| `algolia_search_for_facet_values`               | [Search for facet values](/doc/rest-api/search/search-for-facet-values)               |
| **Recommend**                                   |                                                                                       |
| `algolia_recommendations`                       | [Retrieve recommendations](/doc/rest-api/recommend/get-recommendations)               |
| **Analytics**                                   |                                                                                       |
| `algolia_analytics_click_positions`             | [Get click positions](/doc/rest-api/analytics/get-click-positions)                    |
| `algolia_analytics_no_click_rate`               | [Get no click rate](/doc/rest-api/analytics/get-no-click-rate)                        |
| `algolia_analytics_top_searches_without_clicks` | [Get searches with no clicks](/doc/rest-api/analytics/get-searches-no-clicks)         |
| `algolia_analytics_top_filters`                 | [Get top filter attributes](/doc/rest-api/analytics/get-top-filter-attributes)        |
| `algolia_analytics_top_filters_no_results`      | [Get top filters with no results](/doc/rest-api/analytics/get-top-filters-no-results) |
| `algolia_analytics_no_results_rate`             | [Get no results rate](/doc/rest-api/analytics/get-no-results-rate)                    |
| `algolia_analytics_number_of_searches`          | [Get searches count](/doc/rest-api/analytics/get-searches-count)                      |
| `algolia_analytics_searches_no_results`         | [Get searches with no results](/doc/rest-api/analytics/get-searches-no-results)       |
| `algolia_analytics_top_searches`                | [Get top searches](/doc/rest-api/analytics/get-top-searches)                          |
| `algolia_analytics_top_search_results`          | [Get top hits](/doc/rest-api/analytics/get-top-hits)                                  |
| `algolia_analytics_number_of_users`             | [Get users count](/doc/rest-api/analytics/get-users-count)                            |
| `algolia_analytics_top_countries`               | [Get top countries](/doc/rest-api/analytics/get-top-countries)                        |

ChatGPT requires two extra tools (see [OpenAI MCP docs](https://platform.openai.com/docs/mcp#create-an-mcp-server)):

| Tool     | API                                                         |
| -------- | ----------------------------------------------------------- |
| `search` | [Search an index](/doc/rest-api/search/search-single-index) |
| `fetch`  | [Retrieve a record](/doc/rest-api/search/get-object)        |

## Guidelines

* **Authentication required**. Productivity MCP uses your Algolia identity and permissions.
* **Read-only access**. Use it for analysis and exploration. Data changes still happen in the dashboard or APIs.
* **Data scope**. Anything you can access in Algolia is available to the MCP, so treat prompts and outputs as internal data.