How to Configure Windsurf Bigquery Audit Log
To set up a windsurf bigquery audit log, turn on cloud data logs for your project. Then give your AI coding agent a narrow database role. Cognition renamed Windsurf to Devin Desktop on June 2, 2026. The app keeps the same IDE and features under the new name. Most teams let the agent use the local developer account. If that role is too broad, the agent can read or change sensitive tables. How the windsurf bigquery audit log works Windsurf, now known as Devin Desktop, uses an agent that runs te
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 windsurf bigquery audit log, turn on cloud data logs for your project. Then give your AI coding agent a narrow database role. Cognition renamed Windsurf to Devin Desktop on June 2, 2026. The app keeps the same IDE and features under the new name. Most teams let the agent use the local developer account. If that role is too broad, the agent can read or change sensitive tables.
How the windsurf bigquery audit log works
Windsurf, now known as Devin Desktop, uses an agent that runs terminal commands with user permission. It opens a dedicated shell that loads the local shell settings. Because of this, all local aliases and variables are ready for the agent to use. The legacy Cascade agent also talks to local MCP servers to reach outside data systems.
When the agent runs a query through a CLI tool or MCP server, it uses local keys. BigQuery records these actions in its cloud logs based on the active user or service account. To track what the agent does, you must turn on data access logs in your cloud project. These logs capture the SQL text, the user identity, and the time of each job.
A good setup uses a separate service account just for the AI agent. When you give the agent its own identity, its queries stand out in the logs. Otherwise, agent queries look just like human queries from the same workstation. You can then route those cloud log entries to a central table for review.
What native logging solves and what it misses
Cloud audit logs are great for looking into past events. They tell your security team who logged in and which tables they touched. If an agent runs a bad query, you can find the record after it happens. Splitting human and agent accounts also helps you spot strange patterns in your log tables.
Still, native cloud roles and logs have a big operational gap. Cloud permissions only check who logs in and which dataset they can reach. They do not inspect what individual statements or commands do in flight. Once an agent has read rights on a table, it can pull every column and row.
Audit Log Integrity + BigQuery IAM: architecture patterns and best practices
Continue reading? Get the full guide.
Free. No spam. Unsubscribe anytime.
Also, a log entry only shows up after the query finishes. It cannot stop a bad drop command or a huge table scan before it runs. Native logs also do not mask sensitive fields in query results. If a table holds emails, phone numbers, card numbers, or national IDs, the agent receives them. Because the agent loads the local shell settings, any broad key on the laptop is at risk.
Steps to tighten local agent database permissions
Start by removing broad project editor roles from your daily developer laptops. Create a read-only role that only covers the specific views the agent needs for coding tasks. Keep sensitive tables with customer emails or card numbers in a separate dataset. Do not grant the local shell account any rights to that sensitive dataset.
Next, review the local MCP servers and shell scripts that the agent can run. Make sure your team does not store long-lived admin keys in plain shell profiles. Remember that the agent terminal loads the user shell profile on startup. Any secret stored in those files is open to the agent when it runs commands. Check your cloud logs each week to verify that the agent only queries approved tables.
Check database and API queries on the wire
Identity tools, roles, and network rules decide who opens a session. Setup alone cannot check what each query does or mask sensitive values in a response once a session starts. When workflows touch Postgres, MySQL, SQL Server, MongoDB, ClickHouse, Kubernetes APIs, SSH, or internal HTTP APIs, hoop.dev sits inline in the wire protocol beside those services as an open-source proxy, a sidecar for your Postgres or API. On every query an AI agent or engineer runs against production, hoop.dev decides to pass, mask, block, or hold it for human review.
Your login details and client tools stay as they are while hoop.dev guardrails refuse bad deletes or drops before they run, and hoop.dev data masking rewrites matched sensitive values in responses in memory. Furthermore, hoop.dev holds SOC 2 Type II, is MIT licensed, and writes every query and decision to an audit trail that generates evidence for HIPAA, GDPR, and PCI DSS. See the source on GitHub.
Frequently asked questions
Why do Windsurf docs point to Devin Desktop?
On June 2, 2026, Cognition renamed Windsurf to Devin Desktop. It has the same IDE and features, and the old web links now go to the new site.
How does the agent reach BigQuery from a laptop?
The agent runs terminal commands with user permission in a shell that loads local variables. It can also use the legacy Cascade agent to talk to configured MCP servers.
Can BigQuery logs stop an agent from reading sensitive columns?
No, audit logs only record queries after they run. To protect emails or card numbers in flight, you need query inspection or field masking rewriting results.
Related guides
In this post