Blog

Connect ILC MCP to Fishbowl and Gemini

Israel LopezOctober 28, 20263 min

ILC MCP connects Google’s Gemini to your live Fishbowl database so you can ask about your inventory, orders, or customers in plain English and get answers from the actual numbers. This walks through the connect, step by step.

New to this? Start with What is an MCP Query? and When not to use an MCP.

Before you start

You’ll need three things:

  • Your MCP URL. ILC hosts the server and sends you your endpoint. In the steps below it looks like https://yourmcp.example.com/mcp — use the real one we send you.
  • Your own Fishbowl login. You sign in as yourself. It’s never a shared key, and your Fishbowl permissions carry over.
  • A Gemini account.

Connect it

  1. Open Connected Apps. In Gemini, open Settings → Connected Apps.
  2. Add a custom app. Choose Add a custom app.
  3. Point it at ILC. Enter your MCP server URL and click Next.
  4. Link your account. Approve the trust / account-linking prompt and sign in with your own Fishbowl user over OAuth. You’re authorizing read-only access — sql:read and docs:read. Nothing can write.
  5. Ask. Start a chat and ask in plain English — “how many customers do we have?” — and the MCP writes the SQL and answers live.

Gemini’s menu labels move around as Google updates the product, so if Connected Apps or Add a custom app sits somewhere slightly different in your account, look for the custom-app / MCP server option under Settings. The endpoint and sign-in are the same wherever it lives. For Google’s own version of these steps, see their help page on connecting custom apps for Gemini Apps.

A few Gemini eligibility limits

Custom apps in the consumer Gemini app come with conditions Google sets, not ILC:

  • You add the server on the web (not the mobile app), though you can use it in both afterward.
  • It’s currently limited to personal Google accounts — not work or school accounts — for users 18 or older in the US, with Keep Activity turned on.

None of this applies to the ILC MCP endpoint itself; it’s about where Gemini’s custom-app feature is available today.

What’s protected

  • Read-only. The read tier grants sql:read and docs:read only. Gemini can look, not change.
  • Your login, your permissions. You authenticate as yourself through a one-time OAuth sign-in. What you can see in Fishbowl is what the AI can see.
  • Every query logged. Each call is recorded, so there’s always a record of what was asked.

Read-only vs. read-write

ILC MCP comes in a read-only tier and a read-write tier. The steps above use read-only. Read-write grants scopes that can change Fishbowl data, so treat it with the same care as any account that can edit your system.

Prefer a different tool? We have guides for ChatGPT, Claude.ai, Claude Code, and OpenCode.