Model tokens के बिना Repo understanding।
Codna symbols, imports, call paths, tests, और dependencies को एक live graph में parse करता है जिसे agent query कर सकता है।
Symbols mappedevery symbol
Languages250+ supported
LLM खर्च0 tokens
उत्पाद
Map वाला coding agent।
Codna पहला prompt खर्च होने से पहले आपके codebase की संरचना समझ लेता है। Agents तेज़ fix करते हैं क्योंकि वे अनुमान लगाना बंद कर देते हैं।
उत्पाद
Agent को evidence मिलता है, पूरा repo नहीं।
Files को context window में dump करने के बजाय, Codna एक compact bundle बनाता है: suspect files, call chain, failing test, और risk map।
bundle: failing_test: checkout.spec.ts suspect_files: 4 call_paths: 7 estimated_context: ~600 tokens
Core capabilities
Issue से verified pull request तक।
1
Triage
किसी भी local path या git URL को milliseconds में समझें और देखें कि बदलाव कहाँ ज़रूरी है।
2
Fix
Root cause, confidence score, और regression-risk estimate के साथ एक patch generate करें।
3
Review
Blast radius, touched tests, और API impact के साथ generated changes review करें।
4
PR
Evidence attached के साथ verified fix pull requests खोलने के लिए GitHub App का उपयोग करें।
Distribution
Three ways to run Codna.
CLIRun codna fix in any repo, CI job, or container.
MCP serverGive Cursor and Claude codebase understanding as a local tool.
GitHub AppTriage issues and open verified fix pull requests.
Frequently asked
A deterministic engine builds a dependency and blast-radius graph in about 60ms, using zero LLM tokens. That graph produces a focused ~600-token evidence bundle — 162x less context than reading the repository — so the AI agent works only on what matters.
Every fix is verified by your own test suite before it ships. Nothing merges until your tests pass.
Codna supports 250+ languages, and has mapped 130 repositories in 9.2 seconds for zero tokens. If your project has tests, Codna can work with it.
In head-to-head testing across 87 tasks, Codna used 5× fewer tokens than Cursor and ran 1.7× faster, with every fix verified by the project's own tests (87/87). Both agents were measured on the same tasks.
Codna ships as a CLI, an MCP server that works inside Cursor and Claude, and a native GitHub App that opens verified fix pull requests directly in your repo.
No. You can self-host Codna, bring your own API key, and egress is fail-closed. Your code is never used for training.
