Profile has an Activity tab that records what happened on the account. It is the answer to 'where did the credits go' and, in a shared workspace, to 'who exported that'.
What it shows
- Reveals, with the contact and the cost.
- Exports, with the job number and the delivered row count.
- Verification runs and the checks they consumed.
- Sign-ins and key creation.
- Plan and wallet events, cross-referenced with Profile then Billing.
Reading it usefully
- Reconcile a charge. Match a line under Billing to the action that caused it here. Billing answers what was paid; Activity answers what was bought.
- Attribute agent spend. Create one API key per agent or integration, and the activity trail tells you which one is expensive.
- Confirm nothing unexpected ran. After giving an agent spending scopes, this is the page that proves it behaved.
Cost control that is better than watching
If the activity trail is where you discover overspend, the guardrail is in the wrong place. Prefer instructions that make an agent check the balance before spending, separate keys without spending scopes for read-only work, and the wallet rather than a plan when you want a hard ceiling on total exposure.
Every credit movement is logged. There is no unlogged spending path, in the app, the API, the CLI or MCP.