Kodus

Kodus is an AI code reviewer for engineering teams that want automated pull request feedback and control over where the reviewer runs. Its reviewer, Kody, checks changes in the context of the surrounding code.

• Contextual PR reviews: Kody checks callers, tests and code in linked repositories, then posts inline findings with suggested fixes.
• Team review rules: Teams can write instructions in plain language or import conventions from .cursorrules, CLAUDE.md and AGENTS.md. Kody can also check changes against requirements in Jira, Linear and Notion.
• Feedback memory: Developers can discuss findings in a pull request and save corrections for Kody to use in future reviews.
• Tracked suggestions: Unimplemented suggestions from closed pull requests can become issues, which Kody resolves when it finds a fix in a later pull request.
• Git workflow support: Kodus reviews pull requests on GitHub, GitLab, Bitbucket, Azure DevOps and Forgejo/Gitea. Reviews can also run through a local CLI or in CI pipelines.

The AGPL-licensed open source core is free to self-host with Docker Compose or Helm on Kubernetes; Kodus Cloud runs as a hosted service. Both use your own model credentials. Kodus works with Claude, GPT, Gemini and Llama, including self-hosted models through OpenAI-compatible endpoints. Teams can choose a model per repository and set a fallback. Self-hosting keeps the review service on your infrastructure, but using an external model provider sends review data to that provider. Self-hosted instances send an anonymous daily usage heartbeat by default, which can be disabled.