Connect Gemini to your company documents

Does Gemini support MCP servers?

Yes, on more than one surface, and they are not equivalent. Gemini CLI has first-class MCP server configuration and connects to Tatara in one config entry. The Gemini app accepts a custom MCP server URL under Spark’s connected apps, gated by Google’s eligibility rules. Gemini Enterprise has its own separate setup on Google Cloud.

How do you connect Gemini to Tatara?

You give the client Tatara’s address and approve the connection once. Tatara serves MCP over streamable HTTP at a single URL, and takes either a browser sign-in or an access token as the credential, so the steps are the same wherever you start.

• ROUTE 01

  Gemini CLI

  One entry in your settings file, with no conditions attached.

  {
    "mcpServers": {
      "tatara": {
        "httpUrl": "https://www.tatara.app/api/mcp"
      }
    }
  }

  1. 1.Add the entry above to ~/.gemini/settings.json, or to a project’s .gemini/settings.json if you want it scoped to one repository.
  2. 2.Start Gemini CLI. Tatara’s tools appear alongside its built-in ones.
  3. 3.To authenticate without an interactive sign-in, add a headers object carrying "Authorization": "Bearer YOUR_ACCESS_TOKEN", with a token issued in Tatara under Settings, Access tokens.

  Recent Gemini CLI versions also accept url with "type": "http" in place of httpUrl. Both forms work; httpUrl is the one Google’s current docs show.

• ROUTE 02

  The Gemini app

  Gemini on the web and mobile takes a custom MCP server through Spark, subject to Google’s conditions.

  1. 1.Open Connected Apps in Gemini’s settings.
  2. 2.Add a custom app with the Tatara MCP address, https://www.tatara.app/api/mcp.
  3. 3.Sign in to Tatara and approve the consent screen.

  Google’s screen asks for credentials only when a server cannot register itself. Tatara supports Dynamic Client Registration, so there is nothing to paste under the advanced options.

The full procedure, including issuing and revoking access tokens, lives in the Tatara docs. For Gemini’s own screens, see Google on connecting custom apps to Gemini.

What can Gemini do once it is connected?

It gets twenty-three tools against your brain, in five groups: finding and reading documents, writing them, organising them into folders, curating the tag and property vocabulary, and deleting or restoring things. The tools describe themselves, so a connected agent discovers what it needs when it needs it.

Every agent write is governed. A document needs a type, a title, a description and tags steered to your taxonomy, and a write that does not qualify comes back with what is wrong and how to fix it. The agent corrects itself inside its own loop instead of leaving a half-written document for someone to find later. Every save is stamped with the credential behind it, which is why an agent cannot pass itself off as a person.

Underneath, a brain is plain Markdown with YAML frontmatter: one file per concept, with links forming a graph, and nothing you could not open in an editor. You can also install the open source agent skills with npx skills add tatara-app/tatara-skills, which teach Gemini the workflow up front so its first write usually passes.

What are the limits with Gemini?

MCP support is not the same across assistants, so here is the honest version for this one.

• The Gemini app route is the conditional one, and the conditions are Google’s. It needs Gemini Spark access, a personal Google Account rather than a work or school one, activity history left on, and it is not available in every region. If your company signs in with Google Workspace accounts, this route is closed to you today.
• Gemini CLI is a terminal tool. It is the reliable path, but it is a developer’s path, which is worth knowing before you plan to put a non-technical team on Gemini specifically.
• Gemini Enterprise connects custom MCP servers through Google Cloud rather than through either route above. The Tatara address is the same; the setup is not, and it is not covered here.
• Every Tatara connection carries read and write access. There is no read-only credential.

Gemini and your company brain

What does it cost to connect Gemini?

Nothing to start. The free plan holds 500 documents and does not ask for a card. Agents connect to it over MCP and can read your brain and make governed writes to it. Paid plans add headroom, and the pricing page has the current limits.