Open source Git extension

Know which AI wrote each line.

Autter CLI turns AI code into traceable Git history, so new teams can trust what shipped from the first commit.

autter blame
git native

Codex

Claude

Cursor

$ git commit[hooks-doctor 0afe44b2] wsl compat check2 files changed, 81 insertions(+), 3 deletions(-) you  ██░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ ai     6%                                  94% $ autter blame src/log_fmt/authorship_log.rscb832b7 (Sagnik Ghosh  133) pub fn execute_diff(cb832b7 (Sagnik Ghosh  134)     repo: &Repository,fe2c4c8 (claude        138)     // Resolve commitsfe2c4c8 (claude        139)     let (from, to) = match spec {

The Autter standard already has support across the agent tools developers use every day.

Claude Code
Codex
Cursor
GitHub Copilot
Windsurf

A record agents cannot explain away.

Autter records explicit line-level attribution as the code is written, then carries that context through normal Git history.

No workflow change

Prompt, edit, commit, and push exactly as you do today. Autter attaches attribution in the background.

Local-first by default

Use the open source CLI fully offline, with no account required and no prompt upload required.

No git hooks

Autter does not wrap Git or rely on per-repo hooks, so normal Git commands stay fast.

Git-native record

Line attribution is stored with Git Notes, giving teams a durable record across common Git operations.

Install once. Keep committing.

Start local-only, then connect to the Autter platform when your team wants dashboards, prompt search, and audit history.

Read CLI docs

Install

Mac, Linux, and Windows through WSL.

curl -sSL https://autter.dev/install.sh | bash

Attribute

A git blame companion that shows human and AI authorship.

autter blame src/log_fmt/authorship_log.rs

Measure

Summarize accepted AI lines, human edits, tools, and models.

autter stats --json

Onboard

Choose local-only mode or connect to the Autter platform.

autter onboard

Add AI attribution to your next commit.

Install the CLI, choose local-only mode, and see AI authorship without changing how your team writes code.

Install CLI
Capt. Patch

Capt. Autter Patch

Online now

I've seen a lot of codebases. Most teams find out they needed Autter after a bad deploy. What does your PR review process look like right now?

Powered by Autter AI