What your git log does not record about AI help

A developer audited 569 commits across two AI-assisted research repositories and released Git Provenance Auditor, a tool that scans git history for AI assistance trailers such as Co-Authored-By lines from Claude Code and Cursor and tracks revisions to agent working files like agent-notes. The audit, detailed in the paper "What Git Does Not Record" (arXiv:2610.28655), found the trailers capture only part of the trail: in one repository 66 of 121 commits carrying any AI trailer used the standard Co-Authored-By line, while in the other 39 commits revised the agent-notes file but only 1 carried the trailer. The developer warns that citing commit logs as provenance for AI-assisted work leaves most of the trail missing, a gap that matters for audits and for evaluations built on agent-written code.

I audited 569 commits across two of my own AI-assisted research repositories and built a small tool to run the same audit on any public repo. The paper is "What Git Does Not Record" arXiv:2610.28655, NeurIPS 2026 AI-Native Academia Workshop . The tool, Git Provenance Auditor, scans git history for AI assistance trailers for example Co-Authored-By lines from Claude Code and Cursor , counts how often agent working files such as agent-notes get revised, and shows how much of the agent's work the log actually records. The uncomfortable finding: the trailers cover only part of it. In one repository, 66 of 121 commits carrying any AI trailer use the standard Co-Authored-By line. In the other, 39 commits revise the agent-notes file and only 1 carries the trailer. If you cite commit logs as provenance for AI-assisted work, most of the trail is missing. That matters for audits, for evaluations built on agent-written code, and for anyone reconstructing what an agent actually did after the fact. Tool: https://github.com/ktwu01/git-provenance-auditor https://github.com/ktwu01/git-provenance-auditor Paper: https://arxiv.org/abs/2610.28655 https://arxiv.org/abs/2610.28655 Feedback welcome, especially from people maintaining agent-written codebases: what do you wish your git history recorded?