Greptile is a pull request review tool. Codna reviews every pull request with severity, confidence and a clear verdict, and opens the fix when you ask.
The problem
Codna review posts inline findings with severity, category and confidence, approves a clean diff, and never requests changes. Reply @codna fix on a finding, with write access, and Codna opens a pull request with the fix stacked on your branch.
How Codna fixes it
Findings carry severity (high, medium, low), a category (correctness, security, performance) and a confidence score, with a suggested change inline. At most ten findings at 0.75 confidence or higher. Dependency claims are checked against npm and PyPI before posting.
A clean diff at medium and high gets an approval that counts toward branch rules. Otherwise Codna comments. When a required check is red, the review says so next to the verdict.
Reply @codna fix on a Codna finding and Codna opens a pull request stacked on the branch. With --open-pr, or through the GitHub App, the pull request states the issue, the root cause, the symbols touched and a confidence score, and asks for review before merging. Codna never merges.
codna review . --pr 42 --post # on the PR: reply "@codna fix" on a finding
What you get
A codna review check run per job: success with no findings, neutral with findings, failure only when you turn on blocking.
By default Codna reviews the commits since its last review. Pass --full for the whole pull request.
@codna fix as a reply opens the pull request. Forks get the suggested change instead.
The proof
The only measured head-to-head Codna publishes is against Cursor: 87 matched bug-fix cases on identical checkouts, 5× fewer tokens, 1.7× faster, 87 of 87 verified, about $0.02 per fix. Everything else on this page describes Codna's own behaviour, not Greptile's internals.
Codna reviews pull requests and also fixes them. Run both if you like; Codna's check is named codna review and its approval counts toward branch rules.
Findings below 0.75 confidence are dropped, the list is capped at ten, duplicates are deduplicated across pushes, and findings that cannot be anchored to a line are demoted to the summary.
Yes. Codna is a CLI, an MCP server, a GitHub Action and a GitHub App. Add it where it helps and keep the rest of your workflow.
Bring a key from Anthropic, OpenAI, Google Gemini, Groq, Mistral, OpenRouter or xAI, stored in your OS keychain with codna key set, or sign in and use the managed allowance. The repository map is built from import patterns, so it is not tied to one language. On-device recall covers Python, JavaScript, TypeScript, Go, Rust, Java, C, C++, C#, PHP and Ruby. Your own test command verifies the fix.
Understanding runs on your machine and spends no tokens. Only the evidence bundle or the diff reaches your model provider, with your key from the OS keychain. Set privacy.egress to fail-closed in codna.yaml and Codna runs your tests only under kernel-level network denial. Secret redaction is always on. On 87 matched bug-fix cases against Cursor, Codna averaged 16,159 total tokens and about $0.02 of model spend per verified fix, 5× fewer tokens and 1.7× faster. With your own key you pay your provider directly.
Related
