ai-ecoverse/slicc-bench

SLICC Bench What skills and models change in SLICC, the browser-based agent runtime. Each run gives a task to a SLICC leader's agent, with one model and one set of installed skills. A judge then grades the transcript and screenshots against the task's rubric. The runner is packages/bench, and this dataset is its published output. Task format: browser-use's BU Bench V2 format. Each task is { id, task, rubric, weights }, with weights summing to 100. Judge: browser-use's findings… See the full description on the dataset page: https://huggingface.co/datasets/ai-ecoverse/slicc-bench.

<iframe src="https://silytics.si/embed/item/3023b417-b6f8-484b-b622-b7743b47df5b" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.