Reproduce “From Delegation to Action Approval: An Exploratory Study of Developers' Practices in Collaborating with AI Coding Agents” 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

Yinan Wu, Dominik Wermke, Ze Shi Li, Kathryn Thomasset Stolee, Bowen Xu

The authors examine how developers supervise coding agents when an agent requests permission to run a command or access a file. They distinguish reviewing generated code from approving actions before they occur, using an exploratory cross-sectional survey of 203 agent users. Careful code reviewers did not reliably report action approval; the result reflects self-reports, not observed behavior or all developers.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.