Install
CLI को locally install करें। Codna आपकी machine, आपके cloud, या self-hosted infrastructure पर चल सकता है।
pip install codna codna --version
Docs
Codna को एक command में install करें।
CLI-first, MCP-ready, और native GitHub App compatible। Local repo understanding से शुरू करें, फिर fixes और PRs जोड़ें।
Quickstart
पहले, एक repo समझें। फिर Codna को एक specific issue fix करने के लिए कहें।
codna triage . codna fix . --issue "the checkout test is failing"
CLI
Core commands जानबूझकर छोटे और scriptable हैं।
codna triage . --json codna fix . --issue "..." --tests codna fix https://github.com/org/repo --issue "..."
MCP server
Codna को Cursor या Claude में एक line में जोड़ें। आपका agent Codna से symbol maps, dependencies, test impact, और focused evidence bundles मांग सकता है।
codna mcp start --repo .
Native GitHub App
GitHub App install करें ताकि Codna issues triage कर सके, root-cause evidence के साथ comment कर सके, और fix pull requests खोल सके।
# App flow Install codna → select repos → enable fix PRs
Configuration
Bring your own model key, or use Codna's managed model — it's the same Codna either way.
model: provider: openai key: env:CODNA_MODEL_KEY privacy: egress: fail-closed redact_secrets: true
FAQ
क्या Codna मेरे code पर train करता है? नहीं। Codna को customer code पर कोई training न हो, इसके लिए डिज़ाइन किया गया है।
क्या मैं self-host कर सकता हूं? हाँ। Self-hosting हमेशा के लिए उपलब्ध है, Enterprise के लिए supported on-prem और air-gapped options के साथ।
Frequently asked
Install with a single command from your terminal. The codna cli runs locally, so you bring your own API key and your code never leaves your machine unless you choose otherwise.
The MCP server for coding integrates natively with Cursor and Claude. Add it to your editor's MCP config and Codna's graph engine becomes available to your AI assistant without any extra steps.
The GitHub App monitors your repository and opens pull requests with verified fixes automatically. Every fix is validated by your own test suite before the PR is created.
No. Codna is self-hostable, and egress is fail-closed by default. You supply your own API key, and the system is designed so your code is never used for model training.
The deterministic graph engine maps the relevant code first, producing an evidence bundle measured at around 600 tokens. Head-to-head against Cursor, Codna used 5× fewer tokens and ran 1.7× faster, every fix test-verified (87/87).
Codna supports 250+ languages, and has mapped 130 repositories in 9.2 seconds for zero LLM tokens. If your project compiles or resolves dependencies, Codna can graph it.