AiMatch

Kodus

Kodus reviews pull requests with AI, checks changes against team rules and linked requirements, and posts inline feedback. Software engineering teams can use its cloud service or self-host the open-source core with their own model keys.

One of 61 tools in Coding

Visit Kodus Compare alternatives Own this tool? Claim it
Kodus screenshot

Who it's for

  • Engineering teams using GitHub, GitLab, Bitbucket, Azure DevOps, or Forgejo/Gitea that want AI feedback directly in pull requests.
  • Teams with established coding rules or requirements in .cursorrules, CLAUDE.md, AGENTS.md, Jira, Linear, or Notion that want reviews checked against them.
  • Platform and security teams that want to deploy code review on their own infrastructure and choose their model provider.

Not the right fit if…

  • Teams that require GitHub Enterprise Server support outside a beta release: the page lists that integration as beta.

How it fits your workflow

Connect a Git provider and configure a model provider using your own API key. Kodus reviews pull request changes and posts inline comments; pushes, rebases, and force-pushes can trigger reviews.

Configure plain-language rules or import existing rule files. Kody can also check changes against requirements retrieved from Jira, Linear, and Notion, and use context from linked repositories.

Developers can ask Kody about a flagged issue or correct its feedback. Kodus can offer to save a correction as a memory for later reviews; unimplemented suggestions can be turned into issues.

Pricing

Free: Free community edition uses your model keys. Model usage can incur separate charges.

Free plan

$0

  • · Free community edition uses your model keys. Model usage can incur separate charges.

Paid plans

Price not published

  • · See the provider pricing page for current rates, billing periods, and usage limits.

Prices checked on Sep 7, 2026 from the vendor's site. They can change; confirm before you buy.

Key features

Model-agnostic AI reviews
Supports multiple LLM providers (Claude, GPT-4, Gemini, Llama, any OpenAI-compatible endpoint) so teams choose their preferred model.
Open source & self-hostable
Kodus is open source and can run on your infrastructure with self-hosted runners to keep IP inside your environment.
Zero-markup LLM costs
Kodus charges no markup on LLM API calls; you pay the model provider directly while Kodus charges for the platform subscription.
Pull request level integration
Integrates with GitHub, GitLab, Bitbucket, and Azure DevOps to read diffs, post inline comments, and respect existing review workflows.
Custom review rules
Teams define review standards in plain language; Kodus detects rule files from Cursor, Copilot, Claude and other tools and enforces custom rules.
Business-rule validation
Pulls requirements from Jira, Linear, and Notion and checks PRs against those requirements, flagging violations.

Works with

  • Git providers
  • Issue and requirements tools
  • LLM providers
  • Rule sources

Limitations to know

  • Unlimited PR reviews with your own key are still subject to the rate limits imposed by the selected LLM provider.
  • GitHub Enterprise Server support is listed as beta.

Alternatives to consider

  • CodeRabbit AI Code Reviews

    Choose it if you want a review platform that also integrates into IDEs and CLIs, as well as pull requests.

  • Squire

    Choose it if you also want AI-written pull request descriptions and an AI chat colleague alongside code review and team-rule enforcement.

Compare Kodus side by side

Getting started

For self-hosting, deploy on a VM with Docker Compose or on Kubernetes with Helm. A Git-provider connection and model-provider keys are needed to configure reviews.

  1. Start the cloud trial or choose the free self-hosted Community plan.
  2. For self-hosting, clone the repository and deploy with Docker Compose or Helm.
  3. Connect a supported Git provider and configure your model keys.
  4. Add review rules and optionally connect Jira, Linear, or Notion for requirement checks.

Frequently asked questions

Which AI models are supported?

Kodus is model agnostic. You can use Claude, GPT-4, Gemini, Llama, or any OpenAI-compatible endpoint.

Can I restrict the permissions Kodus uses?

Yes, you have full control over the permissions you grant. Kodus operates with the minimum access required to keep your code secure.

Will I be charged for all developers in my organization?

No, you decide who is included in the Kodus team and will only be charged for those users.

Do you train your AI model with my code or data?

No, Kodus does not train its models with customer data. Your data is processed securely and is never used to improve or retrain our AI.

What Git providers are supported?

GitHub, GitLab, Bitbucket, and Azure DevOps. Kodus integrates at the pull request level — it reads diffs, posts inline comments, and respects your existing review workflows. Setup takes under 5 minutes.

What does zero markup mean?

Kodus doesn't add any margin on top of LLM API calls. You pay the model provider directly at their listed price; Kodus' revenue comes from the platform subscription.

Do you store my source code?

No, Kodus does not store your source code. All processing happens in real-time, and no part of your repository is saved on our servers.