Pay for the reviewer, not the tokens
Bring your own Claude or Codex key at list price. One flat fee per developer, with no review caps and no markup.
Pilot
- Every feature, on every repository you pick
- Claude, Codex, or both, on your keys
- No review caps
Per active developer
- Billed only for developers whose pull requests we reviewed
- No review caps and no token markup
- Pilot teams hear the price first
You always know who you're paying, and for what
Model usage goes straight to your provider. Our fee depends only on your team size.
- To Anthropic or OpenAI
- Model usage on your own key, at their published rates
- To Kenrow
- A flat fee per active developer each month
- Never
- A markup on tokens, a review cap, or a switch to our keys
Questions
Who counts as an active developer?
Someone who opened at least one pull request we reviewed that month. Reviewers, bots and people who only read reviews don't count.
Do I need both a Claude and a Codex key?
No, one is enough. With both, each part of a review runs on the model that suits it, and if one provider has an outage the other takes over.
What happens if my key stops working?
The review stops and tells you why, on the pull request and in Kenrow. We never switch to our own keys, so model costs never move without you knowing.
How much will the models cost me?
It depends on the size of your pull requests and the models you allow. We're measuring it during the pilot and will publish typical figures before pricing starts.
Do you mark up model usage?
No. Anthropic or OpenAI bill you directly, at their published rates, and we add nothing on top.