# CLASHarness

> Google Research benchmark and generator of 600 verifiable multi-agent conflict tasks in computer use: stale sub-agent reports, contradictory instructions, shared locks.

- id: `clasharness`
- category: Information (`information`)
- url: https://github.com/google-research/CLASHarness
- repo: https://github.com/google-research/CLASHarness
- docs: https://arxiv.org/abs/2610.19128
- status: active
- pricing: free
- license: Apache-2.0 AND CC-BY-4.0
- tags: benchmark, multi-agent, computer-use, conflict-resolution, dataset
- related: OSWorld (https://indexagentica.com/entries/osworld/); ThinkingBox (https://indexagentica.com/entries/thinkingbox/)
- sources: https://github.com/google-research/CLASHarness , https://arxiv.org/abs/2610.19128 , https://huggingface.co/datasets/yuxianglai117/CLASHarness
- added: 2026-11-04
- updated: 2026-11-04
- last verified: 2026-11-04
- maintainer: Google Research
- submitted by: agentica-curator
- page: https://indexagentica.com/entries/clasharness/
- json: https://indexagentica.com/api/entries/clasharness.json

Paper: arXiv 2610.19128 (Oct 2026). Hierarchical scenario planning plus an eight-step verified micro-pipeline pre-commits visual ground truth before rendering and compiles hybrid code and semantic graders from executed reference trajectories, so every task is solvable by construction. The benchmark has 600 multi-round tasks run in stateful environments with mock sub-agents, approvals, locks and background daemons. Across 15 frontier models, the authors report that fixing state is much harder than spotting conflicts and that models fail over 58% of closing actions such as releasing locks. Code Apache-2.0 (Python 3.11+); dataset CC-BY-4.0 on Hugging Face.