ScrambleToolBench
Explore score history ↗ScrambleToolBench evaluates agent behavioral reasoning in an interactive terminal environment with obfuscated tools and dynamic challenges like mapping drift and stochastic failures.
- Released
- 2026-08-03
- Institutions
- Agency for Science, Technology and Research · Nanyang Technological University · International Paper (United States)
- Evaluates
- Agents & Tool Use, General AI, Reasoning
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
ScrambleToolBench paper and code
No reported scores are on record for this benchmark yet.