1
Map for zero tokens
Codna builds a dependency and blast-radius graph of the repository from its import patterns. No model call, zero tokens.
Comparison
Codna vs Cursor
On 87 matched bug-fix cases, Codna used 5× fewer tokens and ran 1.7× faster than Cursor. Both verified 87 of 87. Codna also runs inside Cursor as an MCP server.
The problem
The agent should not have to find the bug itself.
Cursor is an editor with an agent. Codna is the understanding layer: it maps the repository deterministically, hands the agent an evidence bundle, and reports root cause and risk. You can use it from the terminal or give it to Cursor's agent over MCP.
How Codna fixes it
How Codna works
2
Hand over evidence, not the repo
The agent receives an evidence bundle scoped to the issue: the suspect files, the call paths, the failing test. Codna prints the raw-to-bundle token size on every run.
3
Report, gate, open the PR
Every fix reports root cause, confidence, blast radius and regression risk, and passes a risk gate before it is applied or a pull request opens. 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.
pip install "codna[mcp]" codna mcp install --client cursor
What you get
What you get
Measured: 5× fewer tokens, 1.7× faster
16,159 vs 80,971 average tokens per fix; 13.39 s vs 22.57 s average wall time; 87 of 87 verified for both; about $0.02 of model spend per Codna fix.
A deterministic map
Codna builds a dependency and blast-radius graph of the repository from its import patterns. No model call, zero tokens.
It works inside Cursor
codna mcp install --client cursor adds five tools to Cursor: codna_triage, codna_fix, codna_secure, codna_recall and codna_report_bug.
The proof
Fewer tokens. Faster. Verified.
Frequently asked
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 Cursor's internals.
No. Keep Cursor as your editor and add Codna as an MCP server, or run codna fix from the terminal. The measured comparison is about how the fix is found and produced, not about the editor.
On 87 matched bug-fix cases on identical checkouts: 16,159 vs 80,971 average total tokens per fix (5×), 13.39 s vs 22.57 s average wall time (1.7×), 87 of 87 verified for both, and about $0.02 of model spend per Codna fix.
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
