Reproduce “Agent Contracts: A Formal Framework for Resource-Bounded Autonomous AI Systems” 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
Qing Ye, Jing Tan
The authors propose Agent Contracts, executable agreements that specify an agent’s task, capabilities, resource and time limits, success criteria, and termination conditions. An external monitor tracks execution and constrains delegated budgets, though it can halt only after an LLM call returns. In a 70-problem code-review test, the bundled contracted setup reduced mean token use 90%, without establishing equivalent task success.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.