Track AI Code all the way to production

Observability for your software factory.

Read docs

Tracks code from every major coding agent

We're building the observability, governance and harness engineering stack that helps your team get more done with AI.
Improve Agent Autonomy/Measure ROI/Save Tokens/Ship maintainable, high quality code

Accurate AI-attribution, directly in Git

A Git extension that tracks the AI-code through every commit, merge, cherry-pick, stash/pop, rebase, squash, etc.

No workflow changes

Just prompt and commit

Install the open source Git extension and work as usual. Every line of AI code will be tracked through the entire SDLC. Zero-overhead, no workflow changes, and no git hooks.

AI-Attribution for every line

AI Blame

Git AI links each line of AI code to the agent, model, and prompt that generated it. Attribution survives rebases, merges, stash/pops, squashes, restets, cherry-picks, etc.

Trace the full Lifecycle of AI-Code

Our traces track the path of AI-code through the entire SDLC — see how much makes it to production, how often it gets reworked, and where your time and tokens are spent.

% AI Merged to Production

78%

of production code is AI-generated

Daily Token Spend

$3.1k

across all agents & developers

Generated : Production

45:1

lines generated for every 1 shipped

very high

AI-Code Rework

20%

reworked within a week of shipping

higher than average

Track AI's impact on every PR

Pull requestsLast 30 days · main

Pull request% AICostAuthorAgentModel

Rewrite rewrite-ops attribution: object-based core, trace2 ingestion, fuzzer#1542

88% AI

$42.10

SVsvarlamov

Codex

GPT-5.5

Add idempotency keys to webhook retries#1540

81% AI

$24.60

LTLois Tam

Claude Code

Opus 4.8

Fix auth token refresh race condition#1539

74% AI

$13.05

MCMarcus Chen

Cursor

Sonnet 4.6

Migrate billing cron to durable queue#1536

90% AI

$48.20

SVsvarlamov

Codex

GPT-5.5

Add Codex hooks.json install option#1534

71% AI

$11.40

LTLois Tam

Claude Code

Opus 4.8

Watch the demo

For your team

Git AI For Enterprise

Rollout Git AI to every developer via MDM, track usage, quality, and impact of AI-code, and drive improvements in agent autonomy.

Book Demo

Improve Agent Autonomy

Git AI helps you straighten the path from prompt to production — saving tokens, improving outcomes, and requiring less human attention to complete the same work. Find friction in your traces and scale harness engineering.

promptproduction

31−71%

Human turns

840K−38%

Tokens

211−76%

Tool calls

Autonomy

Measure

Team skills

/extract-component

/add-tests

/migrate-api

+19%

more autonomous

+34%

Code Review accepted

Measure and Improve

Mine agent sessions across your team to build and continuously improve the skills, rules, and context that help agents work more autonomously.

Context

Requirements

Architecture

Decisions

Compound Context

Link the important context from agent sessions to the code it produced, so future agents understand the requirements and architecture choices behind what they build on top of.

Readiness

+58%

agent autonomy

trailing 90 days

time →

Agent readiness

Replace vibes with data. See which parts of your codebase AI works well in — and which need the most investment at the harness layer.

Get full visibility into your software factory

See where AI-code comes from, how much ships, and what it costs — then turn those signals into agents that get more done.

Frequently Asked Questions

How accurate is Git AI?

Git AI is explicit - agents tell it what code they write, and Git AI tracks that code through every Git operation. It never uses heuristics, filewatchers or AI to “detect” AI code. Because it's integrated directly into Git, attributions are accuratly preserved through rebases, merges, stash/pops, squashes, resets, cherry-picks, and all the other git operations developers run daily.

How does it work?

Coding agents call `git ai checkpoint` whenever they write code or modify files. On commit, Git AI stores line-level attribution in Git Notes, linking each line of AI code to the agent, model, and session that created it. When you squash, merge, reset, rebase, stash, or cherry-pick, Git AI moves and merges those attributions so your AI code stays accurately tracked. It never uses AI or heuristics to “detect” AI code — the agents report exactly which lines they wrote, for the most accurate attribution possible.

Does the agent have to commit for Git AI to attribute the code?

No. Git AI works no matter how you commit — your Git client, the Git CLI, and your own Git aliases are all supported.

Is there a performance impact?

Git AI adds zero-overhead to every Git operation. Git AI does not use Git hooks and it does not wrap Git - it's transparent and performant.

Do I have to set up agent hooks?

Nope — Git AI manages each agent's pre/post tool use hooks.

Who uses this?

Hundreds of engineering teams — including many in the Fortune 100 — use Git AI to understand their AI usage and make agents more effective on their codebase.

What's the difference between the open source CLI and the teams version?

The CLI accurately attributes AI code on every commit. The teams version adds a secure prompt store and joins in data from across the SDLC — tying token spend to individual pull requests, calculating % AI by PR, team, and repo, surfacing rework during code review, and even tying incidents back to the AI session that caused them. Self-host it or run it in our cloud.