{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WMCX4AC4YMR2BN64DES6JZYTBV","short_pith_number":"pith:WMCX4AC4","schema_version":"1.0","canonical_sha256":"b3057e005cc323a0b7dc1925e4e7130d5878a9407ee24112b1cd0e450ad20a8d","source":{"kind":"arxiv","id":"2606.22557","version":1},"attestation_state":"computed","paper":{"title":"MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Biqing Qi, Bowen Fu, Bowen Yang, Bowen Zhou, Shijie Wang, Shuang Cheng, Xiaowei Sun, Yibo Zhao, Yikun Fu, Zehao Li, Zhenyu Wu, Zhoumianze Liu, Zichen Ding","submitted_at":"2026-06-21T15:32:57Z","abstract_excerpt":"Computer use agents (CUAs) have advanced rapidly in desktop automation, and a growing number of users deploy CUAs such as OpenClaw on Mac Mini for always-on automation. However, existing benchmarks, including those for macOS, evaluate agents without framework augmentation and rely on binary evaluation. As a result, they fail to capture both the framework capabilities leveraged by modern CUAs and the partial progress on long-horizon, multi-application tasks. We present MacAgentBench, a comprehensive macOS agent benchmark comprising 676 tasks across 25 applications, with nearly 60% involving bot"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.22557","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-21T15:32:57Z","cross_cats_sorted":["cs.CL","cs.HC"],"title_canon_sha256":"57f9bcb1b8852b9f399cd9abc9ef2257c1edfcfaaf1e81d7eb84204548c408c9","abstract_canon_sha256":"08cda773538a5e55d07b05c5b54a5eb020efce52c3d5b7bb38998033a14a77ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:41.535745Z","signature_b64":"MSrSRPjwjuQcfGDn8H8/iW2hMa51lmFhfQnQTGhV4eiXvqhMG82Ve66N3eXluAxDhghp4SHpSbRJcFgy03d3Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b3057e005cc323a0b7dc1925e4e7130d5878a9407ee24112b1cd0e450ad20a8d","last_reissued_at":"2026-06-23T02:13:41.535327Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:41.535327Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.HC"],"primary_cat":"cs.AI","authors_text":"Biqing Qi, Bowen Fu, Bowen Yang, Bowen Zhou, Shijie Wang, Shuang Cheng, Xiaowei Sun, Yibo Zhao, Yikun Fu, Zehao Li, Zhenyu Wu, Zhoumianze Liu, Zichen Ding","submitted_at":"2026-06-21T15:32:57Z","abstract_excerpt":"Computer use agents (CUAs) have advanced rapidly in desktop automation, and a growing number of users deploy CUAs such as OpenClaw on Mac Mini for always-on automation. However, existing benchmarks, including those for macOS, evaluate agents without framework augmentation and rely on binary evaluation. As a result, they fail to capture both the framework capabilities leveraged by modern CUAs and the partial progress on long-horizon, multi-application tasks. We present MacAgentBench, a comprehensive macOS agent benchmark comprising 676 tasks across 25 applications, with nearly 60% involving bot"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22557","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.22557/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.22557","created_at":"2026-06-23T02:13:41.535400+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22557v1","created_at":"2026-06-23T02:13:41.535400+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22557","created_at":"2026-06-23T02:13:41.535400+00:00"},{"alias_kind":"pith_short_12","alias_value":"WMCX4AC4YMR2","created_at":"2026-06-23T02:13:41.535400+00:00"},{"alias_kind":"pith_short_16","alias_value":"WMCX4AC4YMR2BN64","created_at":"2026-06-23T02:13:41.535400+00:00"},{"alias_kind":"pith_short_8","alias_value":"WMCX4AC4","created_at":"2026-06-23T02:13:41.535400+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV","json":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV.json","graph_json":"https://pith.science/api/pith-number/WMCX4AC4YMR2BN64DES6JZYTBV/graph.json","events_json":"https://pith.science/api/pith-number/WMCX4AC4YMR2BN64DES6JZYTBV/events.json","paper":"https://pith.science/paper/WMCX4AC4"},"agent_actions":{"view_html":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV","download_json":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV.json","view_paper":"https://pith.science/paper/WMCX4AC4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22557&json=true","fetch_graph":"https://pith.science/api/pith-number/WMCX4AC4YMR2BN64DES6JZYTBV/graph.json","fetch_events":"https://pith.science/api/pith-number/WMCX4AC4YMR2BN64DES6JZYTBV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV/action/storage_attestation","attest_author":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV/action/author_attestation","sign_citation":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV/action/citation_signature","submit_replication":"https://pith.science/pith/WMCX4AC4YMR2BN64DES6JZYTBV/action/replication_record"}},"created_at":"2026-06-23T02:13:41.535400+00:00","updated_at":"2026-06-23T02:13:41.535400+00:00"}