How to Configure Make Bigquery MCP Server

Published
Length
3 min read

Luis Felipe Dias

All posts

To set up a make bigquery mcp server workflow, link your automation scenario to a hosted Model Context Protocol endpoint. You then give that server a Google Cloud service account key with dataset permissions. Teams often start by assigning broad project roles to get scenarios running fast. This shortcut exposes sensitive tables and allows automated prompts to run expensive or destructive queries. How the make bigquery mcp server workflow operates Make runs automated tasks by linking web apps

Your coding agents run with your access.

Claude Code reaches prod with a developer's credentials, and approval prompts get clicked through. hoop.dev checks every query, masks sensitive data, and holds risky changes for a human.

To set up a make bigquery mcp server workflow, link your automation scenario to a hosted Model Context Protocol endpoint. You then give that server a Google Cloud service account key with dataset permissions. Teams often start by assigning broad project roles to get scenarios running fast. This shortcut exposes sensitive tables and allows automated prompts to run expensive or destructive queries.

How the make bigquery mcp server workflow operates

Make runs automated tasks by linking web apps through visual steps. When you add an AI agent step, it needs a standard way to read data. The Model Context Protocol provides this standard bridge between the AI model and your warehouse. The BigQuery server exposes dataset schemas and query tools to the automation platform.

First, you host the server process on a cloud instance that Make can reach. You give this server a Google Cloud service account credential. Next, you point your Make scenario to the server endpoint. When a trigger fires in Make, the scenario sends a prompt to the model. The model asks the server to list tables or run a SQL query.

The server translates that tool call into a standard BigQuery API request. It runs the query using the attached service account role. BigQuery returns the rows to the server, which sends them back to Make. The scenario then passes those results to the next app in your flow.

What native cloud roles solve and what they miss

Google Cloud IAM roles control which service account can log into the data warehouse. You can limit the account to read-only permissions on a single dataset. You can also set project quotas to cap total daily bytes scanned. These native controls stop outside users from reading your data without a valid key.

Still, static cloud roles only check who logs in at the start. They do not inspect what individual SQL statements do in flight. Every Make scenario using that server shares the exact same database privileges. If a prompt goes wrong, it can pull full tables of customer records.

Native service account keys also sit as static files on your server host. Anyone who gets that key can run queries from outside your automation tool. Also, standard cloud logs only show the shared service account name. You cannot tell which Make scenario or team member triggered a specific query.

BigQuery IAM + Kubernetes API Server Access: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

Why live query checks and data masking matter

Automated workflows move data fast without a human checking each step. When an AI model writes SQL, it often selects every column in a table. That query result might hold emails, phone numbers, card numbers, or national IDs. Sending raw personal data into third-party automation steps creates a major security risk.

Dynamic data masking fixes this problem before the results leave your network. Masking rewrites matched fields in results so sensitive values become redacted text. The automation flow keeps running, and the model only processes safe, scrubbed strings.

Teams also need fine control over which commands run in real time. You might want to block certain table joins or require approval for large scans. Checking each query in flight lets you enforce rules based on the caller identity. You keep your automation fast while keeping tight control over your warehouse data.

Govern statements and responses across databases and APIs

Good config and scoped roles are needed setup. They do not check single queries in flight or mask sensitive fields in returned rows. Across Postgres, MySQL, SQL Server, MongoDB, ClickHouse, Kubernetes APIs, SSH, and internal HTTP endpoints, hoop.dev runs beside each supported service as an open-source proxy, a sidecar for your database or API.

First, hoop.dev guardrails block banned queries before they run. Then hoop.dev data masking rewrites matched sensitive values in responses in memory. The audit trail records every statement and verdict per resource. Furthermore, hoop.dev holds SOC 2 Type II and is open source on GitHub.

Frequently asked questions

How do Make scenarios log in to the MCP server?

Make sends web requests to your hosted server over HTTPS using headers. You verify those requests with an API token and link the server to BigQuery.

Can I block AI agents from dropping BigQuery tables?

Yes, you can assign a read-only data viewer role in Google Cloud. You can also filter live SQL commands to block drop or delete statements.

How do I keep personal data out of Make execution logs?

Apply field masking to your query results before they reach the automation tool. Masking rewrites emails, phone numbers, and card numbers in results during transit.

The query runs. The emails stay hidden.