Skip to main content

What whodunit is

whodunit records which AI agent touched which code, as a plain git trailer on every commit:

AI-Attribution: v=1; status=assisted; method=intersected; agent=claude-code; agent_version=2.1.228; ratio=0.62; model=claude-opus-5; session=a3f9e21c

It exists because the interesting claims about AI-assisted development — how much of the codebase it wrote, what it cost, whether the work came back as rework — are made without evidence. A trailer is evidence: attached to a specific commit, checkable by anyone with the repository, and impossible to add retroactively without rewriting history.

What it is not​

It is not a productivity measurement. This is the most important sentence on the site, so it is near the top rather than in a footnote.

whodunit can tell you that 143 commits carry agent attribution and that they consumed 7.5 billion tokens. It cannot tell you those commits were faster, because assisted and unassisted work are self-selected: people reach for an agent on some kinds of task and not others. A throughput difference between the two groups may be measuring which work was chosen, not what the agent did.

The tool reports cost and cycle time per delivered unit of work, with the denominators visible, and refuses to collapse that into a percentage. See What the numbers mean.

It is not surveillance. It answers "which agent touched this code", not "who wrote this line". No prompt text, no keystrokes, no file contents are recorded — the journal schema has no field that could hold them. Collection is entirely local and makes no network calls; data leaves your machine only if you configure dun sync and run it. See Privacy.

How it works, in three parts​

The trailer is a specification, not an implementation. It is plain git, readable by anything that reads a commit message, and it outlives any one tool that writes it.

The collector (dun) fills the trailer in automatically. It reads the session transcripts your agent already writes for its own purposes, matches them against what you staged, and stamps the commit at commit time. Three agents are supported today, and they do not report the same things — see Agent capabilities before you draw conclusions from an empty panel.

The dashboards are optional. dun sync publishes into an Apache DevLake database, and seven Grafana dashboards read it. Everything before this step works with no network and no server.

Start here​

Built by Navjyot's Lab. Apache-2.0, and the source is on GitHub.