How to Configure Goose Cassandra Audit Log

Published
Length
3 min read

Luis Felipe Dias

All posts

You set up a goose cassandra audit log by turning on native database logging. Then you give the AI agent a distinct database role for its sessions. Teams often link local agents to data stores using shared admin keys. Broad permissions let an automated tool drop tables or read private user rows. A clear log helps you track every action the agent takes. How the goose cassandra audit log works in practice Goose runs as an agent on a developer machine or build server. It reads local files and ru

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.

You set up a goose cassandra audit log by turning on native database logging. Then you give the AI agent a distinct database role for its sessions. Teams often link local agents to data stores using shared admin keys. Broad permissions let an automated tool drop tables or read private user rows. A clear log helps you track every action the agent takes.

How the goose cassandra audit log works in practice

Goose runs as an agent on a developer machine or build server. It reads local files and runs shell commands to finish coding tasks. When a task needs database changes, the agent starts a database session. It uses a local profile with a username and a password.

Cassandra includes a built-in audit logging feature on each cluster node. Admins turn this feature on in the main node settings file. Once active, the database writes every login attempt and query to a local file. You can filter these logs by role name or command type.

To make the goose cassandra audit log useful, create a dedicated role for the agent. Never let the agent share a role with a human developer. When the agent runs a query, the node records that specific role name. Your security team can then ship those node files to a central log store. This setup shows you which queries came from the AI agent.

What native logging solves and where it falls short

Native audit logs tell you what happened after a query finishes. They prove which role ran a command and when the event took place. This history helps teams pass security reviews and debug broken database states. It separates bot traffic from human work during an incident review.

Still, native node logs do not stop bad queries in flight. A database role decides who can log in and which tables they can touch. Once the session starts, the database trusts every valid command from that role. If the agent hallucinates a destructive command, the node runs it right away. The audit log only records the damage after the data is gone.

Audit Log Integrity + Cassandra Role Management: architecture patterns and best practices

Continue reading? Get the full guide.

Free. No spam. Unsubscribe anytime.

Native logs also miss the human identity behind a shared agent setup. If five engineers run the same agent profile, the log shows one role. You cannot tell which engineer prompted the agent to run a specific query. Also, native logs do not mask sensitive fields in query results. If the agent reads a table, raw emails and card numbers flow back to the model.

Why agent sessions need statement-level controls

AI agents work fast and run many commands without getting tired. Static database roles are too blunt to manage this kind of automated traffic. An agent might need to read schema details and update one specific row. Giving that agent full write permissions across a keyspace creates huge risk.

Teams need a way to check each query before the database runs it. If an agent tries to delete a table, the system should pause that command. A human reviewer can then approve or deny the action in real time. At the same time, data masking should rewrite sensitive fields in the results. Rewriting phone numbers and national IDs keeps private data out of model prompts. Pairing real-time checks with your node logs gives you full control over agent work.

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 runs beside those services as an open-source proxy, a sidecar for your Postgres or API. It inspects requests coming in and results going back on the wire.

Your login details and client tools stay as they are while hoop.dev guardrails refuse bad deletes or drops before they run. On responses, hoop.dev data masking rewrites matched sensitive values 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

Does Goose have its own database audit log?

Goose logs its own chat history and tool runs on the local machine. Still, a true database audit trail must live outside the client machine. You need server-side logs so a local user or agent cannot delete the record.

How do I separate human queries from agent queries in Cassandra?

Create a unique database role just for the AI agent to use. Set the native audit log to track all events from that specific role. This keeps automated actions separate from human logins in your log files.

Can native Cassandra logs block a bad query from an agent?

No, native audit logs only record events after or as they happen. To block a dangerous query before it runs, you need inline statement checks. You can also restrict the database role to read-only permissions to stop writes.

The query runs. The emails stay hidden.