Feature
Observability and transparency in ClawAI
How ClawAI shows what a request is spending — a usage dashboard, per-answer routing detail, an audit log and live progress — so usage is never a black box.
All features · Last reviewed:
The usage dashboard, and per-answer routing detail
A usage dashboard shows the allowance consumed today and this month, broken down by model, with the remaining balance in the same units a plan is quoted in. Underneath that, every individual answer carries the model that produced it, why it was chosen, how long it took, how many tokens it used, and what it cost against the allowance — including which model stepped in if the original provider failed mid-request. See what is AI model routing and what is model fallback, both linked below, for how that routing decision itself gets made.
An audit log for logins, plan changes and connector activity
Logins, plan changes, connector activity, memory edits and generated content are each recorded with a timestamp and an actor, so an account’s history is reconstructable rather than only visible in the moment it happened. This is the same audit trail referenced from the memory-and-context and security-and-data-handling pages, linked below, for the actions each of those features writes into it.
Live progress while a model works, and clear limit warnings
While a model works you see the stage it is in, the text as it arrives, its reasoning when the model exposes it, and running token and timing counters — nothing is a silent wait. When a request reaches a plan limit, ClawAI states which limit, how much is left on the other windows, and when it resets, rather than cutting the request off with no explanation.
Questions people ask
- Can I see which model answered a specific message, and why?
- Yes — every answer records the provider and model that produced it, the reasoning behind the routing choice, how long it took, how many tokens it used, and what it cost against your allowance.
- What does the audit log actually record?
- Logins, plan changes, connector activity, memory edits and generated content, each with a timestamp and an actor, so an account’s history can be reconstructed after the fact.
- What happens when I hit a usage limit?
- ClawAI states which specific limit was hit, how much allowance remains on your other usage windows, and when the limit resets — nothing is cut off silently.
Try it rather than take our word for it
A usage dashboard, per-answer routing detail, an audit log and live progress are four real, shipped pieces of the same principle: usage in ClawAI is metered, attributed and visible, not a black box.