{"slug":"amap-ml-weavebench-cua-reproduce","source_name":"amap-ml/weavebench-cua-reproduce","name":"Amap ML/Weavebench Cua Reproduce","description":"Reproduce CUA-Harness experiments on WeaveBench from a GitHub checkout. Use when the user wants an AI coding agent to set up dependencies, download WeaveBench assets, prepare the 120G VM, configure Qwen/Anthropic-compatible APIs, run smoke tests, launch full or subset evaluations, inspect logs, or summarize scores for this repository.","version":1,"lift":{"pass_rate_delta_pts":43.48,"pass_rate_pct":56.5,"total_cases":23,"passed_cases":13,"tokens_delta_pct":65.2,"turns_delta_pct":0,"verdict":"mixed","benchmark_model":"gemini-3.6-flash","grading_method":"judged","completed_at":"2026-08-31T04:21:46.178838+00:00"},"skill_score":0.5652,"benchmark_models":[{"model":"gemini-3.6-flash","headline":true,"delta_pts":43.48,"with_pass_pct":56.5,"without_pass_pct":13,"tokens_delta_pct":65.2,"turns_delta_pct":0,"total_cases":23,"cases_aggregated":17,"verdict":"mixed","never_hurt":false,"completed_at":"2026-08-31T04:21:46.178838+00:00","run_id":"83eead09-cf15-438d-a099-7988fd3d53af","version_number":1,"is_latest_version":true,"gate":null}],"trust":{"skill_safety":"passed","safety_status":"clean","intent_verdict":"safe","content_status":"clean","indexable":true},"license":"MIT","install_count":0,"manifest_hash":"3c81016debfbe5e4c9235aea633d766671078b14f253b89892843cddafda4202","raw_url":"https://app.decimal.ai/s/amap-ml-weavebench-cua-reproduce/SKILL.md","scorecard_url":"https://app.decimal.ai/skills/amap-ml-weavebench-cua-reproduce"}