Reproduce “Agentic process evidence” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Yuval Kainan, Carmit Hershman
The authors propose Agentic Process Evidence (APE), a way to link captured AI-agent session records to the software outcome they helped produce, such as a commit. The evidence uses signed, machine-readable attestations that a later gate can check. APE describes a governance design, not a deployed system or measured security improvement; signatures do not prove capture is complete or truthful.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.