{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QKVGUKIKKKKOJPZKDRNZGUCD5W","short_pith_number":"pith:QKVGUKIK","canonical_record":{"source":{"id":"2407.00132","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-06-28T08:45:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2b4917f4cd27921bbf67499dd8c34b680f0a83149aafa45f1535b15556af7689","abstract_canon_sha256":"7029a77dc052811609ae5d2435d612b280181461e34ab221484e5401f58c7db3"},"schema_version":"1.0"},"canonical_sha256":"82aa6a290a5294e4bf2a1c5b935043eda9804c821a58668f8d118d831871d14d","source":{"kind":"arxiv","id":"2407.00132","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00132","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00132v3","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00132","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"pith_short_12","alias_value":"QKVGUKIKKKKO","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"pith_short_16","alias_value":"QKVGUKIKKKKOJPZK","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"pith_short_8","alias_value":"QKVGUKIK","created_at":"2026-07-05T10:04:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QKVGUKIKKKKOJPZKDRNZGUCD5W","target":"record","payload":{"canonical_record":{"source":{"id":"2407.00132","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-06-28T08:45:02Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2b4917f4cd27921bbf67499dd8c34b680f0a83149aafa45f1535b15556af7689","abstract_canon_sha256":"7029a77dc052811609ae5d2435d612b280181461e34ab221484e5401f58c7db3"},"schema_version":"1.0"},"canonical_sha256":"82aa6a290a5294e4bf2a1c5b935043eda9804c821a58668f8d118d831871d14d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:04:14.252843Z","signature_b64":"wzlkg/llLjHf7GHvvco8v4+o1Pl3n9kLOY4M+xM6EtWdZ3Esj2O3EDlLSLFX3jPbe6ylx57Z3WhwuTL2lIIEBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"82aa6a290a5294e4bf2a1c5b935043eda9804c821a58668f8d118d831871d14d","last_reissued_at":"2026-07-05T10:04:14.252430Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:04:14.252430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.00132","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:04:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0cogbHOzkiK2Nsazh2KUCt/Kb9So0/MYwxJDcUPyg6fmJMz9mEbmmnvtPuFgMx1cMH+Ibmt3NT+x6W4c9hpgAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T12:42:51.054564Z"},"content_sha256":"757d0bc37073a2dbbcd7a8f122b3ee561f564b2858ae6f0f40c7a9c710aa075b","schema_version":"1.0","event_id":"sha256:757d0bc37073a2dbbcd7a8f122b3ee561f564b2858ae6f0f40c7a9c710aa075b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QKVGUKIKKKKOJPZKDRNZGUCD5W","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ShortcutsBench: A Large-Scale Real-world Benchmark for API-based Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Desong Meng, Dongqi Cai, Haiyang Shen, Li Zhang, Mengwei Xu, Sheng Qi, Yue Li, Yun Ma","submitted_at":"2024-06-28T08:45:02Z","abstract_excerpt":"Recent advancements in integrating large language models (LLMs) with application programming interfaces (APIs) have gained significant interest in both academia and industry. Recent work demonstrates that these API-based agents exhibit relatively strong autonomy and planning capabilities. However, their ability to handle multi-dimensional difficulty levels, diverse task types, and real-world demands remains unknown. In this paper, we introduce \\textsc{ShortcutsBench}, a large-scale benchmark for the comprehensive evaluation of API-based agents in solving real-world complex tasks. \\textsc{Short"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00132","kind":"arxiv","version":3},"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/2407.00132/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:04:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LAYofw/5Gcxa25PNBk6l8aKbeV8FP1A/Dq00uPtpdgB9Cgxs6h2x0uvjidgXRgruOpIdAfR/tAF9llOOZmvNDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T12:42:51.055157Z"},"content_sha256":"d5377de052617eb53fbe6861d3718197989482ad2fa2913e0f9cb81e5ad376ad","schema_version":"1.0","event_id":"sha256:d5377de052617eb53fbe6861d3718197989482ad2fa2913e0f9cb81e5ad376ad"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W/bundle.json","state_url":"https://pith.science/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-10T12:42:51Z","links":{"resolver":"https://pith.science/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W","bundle":"https://pith.science/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W/bundle.json","state":"https://pith.science/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QKVGUKIKKKKOJPZKDRNZGUCD5W/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QKVGUKIKKKKOJPZKDRNZGUCD5W","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"7029a77dc052811609ae5d2435d612b280181461e34ab221484e5401f58c7db3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-06-28T08:45:02Z","title_canon_sha256":"2b4917f4cd27921bbf67499dd8c34b680f0a83149aafa45f1535b15556af7689"},"schema_version":"1.0","source":{"id":"2407.00132","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.00132","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"arxiv_version","alias_value":"2407.00132v3","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.00132","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"pith_short_12","alias_value":"QKVGUKIKKKKO","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"pith_short_16","alias_value":"QKVGUKIKKKKOJPZK","created_at":"2026-07-05T10:04:14Z"},{"alias_kind":"pith_short_8","alias_value":"QKVGUKIK","created_at":"2026-07-05T10:04:14Z"}],"graph_snapshots":[{"event_id":"sha256:d5377de052617eb53fbe6861d3718197989482ad2fa2913e0f9cb81e5ad376ad","target":"graph","created_at":"2026-07-05T10:04:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2407.00132/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advancements in integrating large language models (LLMs) with application programming interfaces (APIs) have gained significant interest in both academia and industry. Recent work demonstrates that these API-based agents exhibit relatively strong autonomy and planning capabilities. However, their ability to handle multi-dimensional difficulty levels, diverse task types, and real-world demands remains unknown. In this paper, we introduce \\textsc{ShortcutsBench}, a large-scale benchmark for the comprehensive evaluation of API-based agents in solving real-world complex tasks. \\textsc{Short","authors_text":"Desong Meng, Dongqi Cai, Haiyang Shen, Li Zhang, Mengwei Xu, Sheng Qi, Yue Li, Yun Ma","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-06-28T08:45:02Z","title":"ShortcutsBench: A Large-Scale Real-world Benchmark for API-based Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.00132","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:757d0bc37073a2dbbcd7a8f122b3ee561f564b2858ae6f0f40c7a9c710aa075b","target":"record","created_at":"2026-07-05T10:04:14Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"7029a77dc052811609ae5d2435d612b280181461e34ab221484e5401f58c7db3","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-06-28T08:45:02Z","title_canon_sha256":"2b4917f4cd27921bbf67499dd8c34b680f0a83149aafa45f1535b15556af7689"},"schema_version":"1.0","source":{"id":"2407.00132","kind":"arxiv","version":3}},"canonical_sha256":"82aa6a290a5294e4bf2a1c5b935043eda9804c821a58668f8d118d831871d14d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"82aa6a290a5294e4bf2a1c5b935043eda9804c821a58668f8d118d831871d14d","first_computed_at":"2026-07-05T10:04:14.252430Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:04:14.252430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wzlkg/llLjHf7GHvvco8v4+o1Pl3n9kLOY4M+xM6EtWdZ3Esj2O3EDlLSLFX3jPbe6ylx57Z3WhwuTL2lIIEBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:04:14.252843Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.00132","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:757d0bc37073a2dbbcd7a8f122b3ee561f564b2858ae6f0f40c7a9c710aa075b","sha256:d5377de052617eb53fbe6861d3718197989482ad2fa2913e0f9cb81e5ad376ad"],"state_sha256":"bd3e6c66202f3ad84c90261f941fcbc490b74c0e7845fd40e23635d03c3daaeb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"T/SAvmng2XeGFr20I+kDvkE7fYfL1TiG89dIPflVZo4SlWsCVZRH5s9LIogIP49B4uoRxqDtGMQqiVOcIKIMDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T12:42:51.060676Z","bundle_sha256":"df3cc2f68735a9ff6f9505b372dfec9ca144977afd59a3717e7643b5edeaa1b7"}}