> For the complete documentation index, see [llms.txt](https://docs.siit.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.siit.io/guides/integrations/connecting-siit-mcp-to-google-gemini-enterprise.md).

# Connect the MCP server to Gemini Enterprise

Connect the Siit MCP server to Google Gemini Enterprise: required Siit and GCP roles, the Google Cloud organization policy change for custom MCP endpoints, and the setup steps

<figure><img src="https://2274035325-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FlWFhpq2ia6gdHet0opKx%2Fuploads%2Fgit-blob-a3aba26ffdc28a07e2f363050d8419fff95320ce%2Fundefined-2.png?alt=media" alt=""><figcaption></figcaption></figure>

### Prerequisites & Requirements

Before initiating the installation, ensure you have administrative permissions across both platforms:

* **Siit Environment:** Access to the *Admin Console* with Public API & MCP permissions
* **Google Cloud Platform (GCP):** An active project tied to your Gemini Enterprise licenses with the Discovery Engine Editor (roles/discoveryengine.editor) and Organization Policy Administrator IAM roles.

### Step 1: Modify Google Cloud Organization Policies

By default, Google Cloud applies a restrictive management constraint that restricts custom third-party MCP endpoints. This safety barrier must be bypassed at the organizational or project level before proceeding.

1. Log into the **Google Cloud Console** and navigate to the **Organization Policies** dashboard.
2. Search for the following constraint: constraints/gcp.disableCustomMcpServer (*"Disable custom MCP server connector for Gemini Enterprise"*).
3. Select the policy, click **Edit Policy**, and choose **Customize**.
4. Under Enforcement status, select **Off** to allow custom MCP data stores within your designated ecosystem. Click **Save**.

*Note: Google Cloud policy adjustments can take up to 5 to 10 minutes to propagate globally across your cloud environment.*

### Step 2: Create the Custom MCP Data Store in Gemini

With credentials active, you can now construct the architectural container within Google Cloud that handles your requests.

<figure><img src="https://2274035325-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FlWFhpq2ia6gdHet0opKx%2Fuploads%2Fgit-blob-c3b480a63315005b8fa7e1388ea71b2078099802%2Fundefined-3.png?alt=media" alt="Third party sources on Gemini Enterprise"><figcaption></figcaption></figure>

1. In the Google Cloud Console, navigate to **Gemini Enterprise** (or **Vertex AI Search and Conversation**).
2. In the left-hand navigation pane, select **Data Stores**, then click **+ Create Data Store**.
3. From the available sources list, select the **Custom MCP Server (Preview)** card.
4. Fill out the form fields:

* **Transport Type:** Choose Streamable HTTP (SSE mode).
* **Server URL:** Insert your unique Siit Server URL endpoint.
* **Authentication Protocol:** Select OAuth 2.0 / 2.1.
* **Scopes:** Enter all available scopes.

5. Scroll down to the **Model Routing Description** field. This natural language explanation instructs Gemini's core orchestration layer on exactly when to invoke the Siit connection. Copy and paste this description text:\
   *"Use this server application connector to interact directly with the Siit enterprise IT Service Desk and operational platform. This system handles reading, indexing, creating, updating, or assigning internal IT tickets, customer help requests, managing employee equipment or software provisions, looking up user group memberships, and looking up organizational knowledge base documentation stored within Siit."*
6. Name the data store explicitly (e.g., Siit-Service-Desk-MCP) and click **Create**.

### Step 3: Audit & Enable Siit Functional Tools

For explicit safety control, Gemini imports external capabilities in an inactive state. You must manually unlock the precise functions your workspace requires.

1. Open your newly established Siit-Service-Desk-MCP data store from the console inventory.
2. Click into the **Actions / Tools** control tab.
3. Click **Reload Custom Actions**. This commands Google to query Siit's available endpoints. The interface will populate with operational tools, such as siit\_get\_ticket, siit\_create\_ticket, and siit\_lookup\_hardware.
4. Check the activation box next to each action your team needs access to, and click **Enable Actions**.

### Step 4: User Verification & Sign-on

Once deployed globally by an administrator, individual employees can invoke the tool within their conversational workflow.

1. Instruct an authorized employee to navigate to their main **Gemini Enterprise Chat Workspace**.
2. At the bottom left of the chat window, click the **Data Connections / Extensions** toggle array.
3. Locate **Siit-Service-Desk-MCP** and switch the toggle to **Active (ON)**.
4. The browser will present an automatic OAuth login prompt. The user must sign in securely using their corporate Siit identity credentials to complete the authorization loop.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.siit.io/guides/integrations/connecting-siit-mcp-to-google-gemini-enterprise.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
