BigQuery MCP server
Turn approved BigQuery queries into MCP tools
Expose selected BigQuery analytics workflows through a hosted MCP endpoint with typed parameters and a client-readable contract.
- Published
- 2026-08-31
- Reviewed by
- Rachid
Direct answer
Direct answer
Use Datatape with BigQuery when agents should run known analytics workflows without receiving a general query surface. Google’s open-source MCP Toolbox is a better fit when you want to build and operate the MCP infrastructure yourself.
Keep analytics workflows intentional
BigQuery makes broad exploration possible, but an operational agent often needs only a known metric over a bounded period. A named tool lets the data team own that query while the client supplies only the approved inputs.
Managed and self-hosted are different choices
Google’s MCP Toolbox for Databases supports BigQuery and custom tools as an open-source framework. Choose it for deployment control. Choose Datatape when operating the endpoint and maintaining a cross-source catalog should be handled as a product workflow.
campaign_summary
Summarize approved campaign metrics for a date range.
- start_date
- date · required
- end_date
- date · required
SELECT campaign_name, SUM(spend) AS spend, SUM(conversions) AS conversions
FROM analytics.campaign_daily
WHERE event_date BETWEEN {{ start_date }} AND {{ end_date }}
GROUP BY campaign_nameReturns: Campaign-level spend and conversions for the requested period.
A practical operating model
MCP does not replace database security. Pair the tool contract with controls at the source.
Use a dedicated service account
Grant the connection only the BigQuery roles and datasets required for these tools. Avoid project-wide ownership roles.
Bound expensive scans
Prefer partition filters, governed views, and explicit date ranges. Use BigQuery cost controls independently of the MCP layer.
Publish business definitions
A tool should state the source view and metric meaning clearly enough that reviewers can verify the query against the description.
Decision support
Compare the operating models
| Decision | Primary benefit | Alternative / tradeoff |
|---|---|---|
| Operating model | Hosted, reviewed tool catalog | Open-source framework you deploy and operate |
| Source scope | BigQuery plus databases, HTTP APIs, and other sources | Database-focused toolbox with BigQuery support |
| Choose it when | You want a managed endpoint and fixed workflows | You want infrastructure control and custom deployment |
Client setup
Connect the clients your team already uses
The same hosted endpoint can be registered in multiple compatible clients. OAuth is the default human flow; agent API keys support non-interactive clients where headers are available.
Add the remote server from Settings → Connectors, then complete the browser authorization flow.
- Step 1Open Settings → Connectors and choose Add custom connector.
- Step 2Enter https://<org-id>.mcp.datatape.ai/mcp as the remote MCP server URL.
- Step 3Connect and finish authorization in the browser.
Client settings can change between releases. Verify the transport and authentication flow against the current client documentation.
Evidence
Sources and review
Written by Datatape Editorial. Product and technical claims reviewed by Rachid.
Continue