How to Configure Windsurf Bigquery Approval

Published
Length
3 min read

Luis Felipe Dias

All posts

You set up windsurf bigquery approval by requiring user permission before the agent runs commands. You also pair those local prompts with strict Google Cloud roles. Most teams start by linking their editor to cloud data through local terminal sessions. When cloud roles are too broad, a single approved command can alter production tables. It can also pull sensitive data into the editor context. How windsurf bigquery approval works in the editor On June 2, 2026, Cognition renamed the Windsurf e

Approval Chains & Escalation + BigQuery IAM: the complete guide

Architecture patterns, implementation strategies, and security best practices. Delivered to your inbox.

Free white paper. No spam. Unsubscribe anytime.

You set up windsurf bigquery approval by requiring user permission before the agent runs commands. You also pair those local prompts with strict Google Cloud roles. Most teams start by linking their editor to cloud data through local terminal sessions. When cloud roles are too broad, a single approved command can alter production tables. It can also pull sensitive data into the editor context.

How windsurf bigquery approval works in the editor

On June 2, 2026, Cognition renamed the Windsurf editor to Devin Desktop. The app keeps the same features and uses the Cascade agent to run tasks. The agent can run terminal commands on its own with user permission. It uses a dedicated terminal that loads the user shell setup. On macOS it runs zsh, and on Linux it runs bash. Because it loads the shell setup, all local aliases and variables are active. If a developer logs into Google Cloud locally, the agent inherits that active session. The agent can also reach external tools through Model Context Protocol servers. When the agent wants to query a dataset, it proposes a command or tool call. The developer reviews the prompt inside the editor and clicks to approve it. This step acts as the primary local check before any query runs.

What native cloud roles solve and what they miss

Google Cloud IAM roles control which datasets and tables a user or service account reaches. Admins can assign read-only roles to limit damage from bad queries. Together, local editor prompts and cloud roles handle basic login and role checks. Still, this native setup leaves an operational gap during live work. Cloud IAM decides who can log in, not what individual statements do in flight. If a developer needs write permissions for one task, the agent gets those same permissions. A developer tired of constant prompts might approve a dangerous command by mistake. Also, read-only roles do not stop an agent from reading sensitive columns. Full rows containing emails, phone numbers, card numbers, or national IDs return to the editor. Native cloud logs record the query, but they do not link it to the prompt. Security teams cannot easily see when a human approved a specific agent action.

Why team review beats local self-approval

Local prompts rely on one developer making the right call every time. In a fast coding session, self-approval does not match real production change control. Teams need a second person to check risky queries before they touch live data. They also need data masking that rewrites matched fields in results before the agent reads them. Moving approval out of the local editor gives security teams clear rules for every query.

Approval Chains & Escalation + BigQuery IAM: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

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 sits inline in the protocol beside each supported service as an open-source proxy, a sidecar for your database or API. For every query or command an AI agent or engineer runs against production, hoop.dev decides to pass, mask, block, or hold it for human approval.

First, hoop.dev guardrails block banned queries before they run or hold risky ones for review, and hoop.dev data masking rewrites matched sensitive values in responses in memory. Furthermore, hoop.dev holds SOC 2 Type II and is open source on GitHub.

Frequently asked questions

Did Windsurf change its name to Devin Desktop?

Yes, on June 2, 2026, Cognition announced that Windsurf is becoming Devin Desktop. It keeps the same IDE features, and the old docs links now redirect there.

How does the Cascade agent reach BigQuery from the editor?

The agent runs commands in a dedicated terminal that loads the local shell setup. It can also use configured Model Context Protocol servers to talk to external data systems.

Why are local editor prompts not enough for production data?

Local prompts only ask the current user for permission and use their active cloud credentials. They cannot enforce peer review or rewrite emails and card numbers in query results.

Save the open-source gateway.

hoop is MIT-licensed infrastructure for controlling how AI agents reach production data. Star hoophq/hoop so you can inspect it, deploy it, or share it when your team starts governing agent access.