UniClawBench: Benchmark for Proactive AI Agents on Real-World Tasks
University of Hong Kong
UniClawBench (arXiv 2607.08768) evaluates proactive agents across five capability dimensions — Skill Usage, Exploration, Long-Context Reasoning, Multimodal Understanding, and Cross-Platform Coordination — with 400 bilingual tasks run inside live Docker containers. A closed-loop evaluation uses multiple agent roles to simulate realistic human feedback without leaking grading criteria to the system under test.
Why it matters
Addresses a real gap in agent benchmarking by requiring agents to act proactively in dynamic environments rather than respond to pre-specified task descriptions.
Importance: 2/5
New proactive-agent benchmark with realistic evaluation methodology; fills gap left by existing static benchmarks