A reviewer you can see into.
Run Cawdit as a server and it keeps every review in Postgres: what it found, what got fixed, which rules it read and what each model call cost.
- 2
- 10
- 2
- 1
What it found, and what it cost.
Pull requests reviewed, findings posted, the share that got fixed and the cost of the model calls, per period and repository.
- Findings per day by severity
- Cost per day in US dollars
- The repositories with the most findings and the highest cost
Every review, step by step.
Each run shows its trigger, the commit, the model, the findings it posted and fixed, the tokens and the cost, next to every other run of the same pull request.
- Automatic, CI and command triggers
- A timeline of the runs of one pull request
- Review again from the dashboard
The rules your repository sets.
A repository's page shows the effective .cawdit.toml of its default branch, the rule files Cawdit loads from it and any warnings, so the people who set it up see what the next review will follow.
- Runs, configuration and connection tabs
- The review command ready to copy
- Warnings for unknown keys
Connect once, per organisation.
Connect an organisation and Cawdit installs and repairs its webhooks. Turn reviews on or off per repository, and sign in through the forge or an OpenID Connect provider.
- Webhook sync for organisations and repositories
- Organisation tokens sealed with AES-256-GCM
- SSO through providers such as Zitadel
One binary. Three ways to run it.
The same Go program reviews from a CI step or serves webhooks and a dashboard. The server's only dependency is Postgres. The images are signed with cosign and run as a user without root.
Crow plugin
One review per pipeline
Add a step to the pull request pipeline. The image reviews the pull request, posts the review and exits, with no database to run.
cawdit serve
Webhooks and a dashboard
Takes the forge's webhooks, queues reviews in Postgres, answers commands in threads, and shows runs, findings and cost per repository.
Helm chart
The server on a cluster
Runs cawdit serve on Kubernetes. The chart bundles no database and never creates a Secret, so yours stay where you keep them.
when:
- event: pull_request
skip_clone: true
steps:
review:
image: docker.io/cawdit/cawdit:plugin-main
settings:
model: claude-opus-5-5
verify_model: claude-sonnet-5-5
api_key:
from_secret: anthropic_api_key
forge_token:
from_secret: cawdit_forge_token
fail_on: errorThe quickest start is a pipeline step.
Give the step a model key and a forge token as secrets, and open a pull request. Set dry_run to print the review instead of posting it while you try it out.
Start with one repository.
The quick start reviews your first pull request from a Crow CI step in about fifteen minutes. Docker Compose and the Helm chart run the server when you want the dashboard and commands.
- Forgejo and Arbinel.
- Your model, your key.
- Two passes per finding.
- One binary, one database.
- AGPL-3.0-or-later, every feature.