STAGE-Claw: Automated State-based Agent Benchmarking for Realistic Scenarios

In three linesSTAGE-Claw is an automated framework for building and evaluating AI agents in realistic scenarios. It automatically generates tasks, environments, and state-based metrics. A benchmark of 40 tasks evaluates 11 frontier models on tool-call reliability and failure patterns.

Summary generated by Claude — human-verified