{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IYJKHMX7J73ZG5LO4MMAPPN655","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":"a41d80bf262fe501d30e568d92f73f98eb919970804ba764aba2c6acf257fb86","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-17T06:37:04Z","title_canon_sha256":"29507e0cd5028ab233f6d828aa1c66393e13166fb034674024c2886cf2f0971d"},"schema_version":"1.0","source":{"id":"2605.17281","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.17281","created_at":"2026-05-20T00:03:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.17281v1","created_at":"2026-05-20T00:03:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.17281","created_at":"2026-05-20T00:03:49Z"},{"alias_kind":"pith_short_12","alias_value":"IYJKHMX7J73Z","created_at":"2026-05-20T00:03:49Z"},{"alias_kind":"pith_short_16","alias_value":"IYJKHMX7J73ZG5LO","created_at":"2026-05-20T00:03:49Z"},{"alias_kind":"pith_short_8","alias_value":"IYJKHMX7","created_at":"2026-05-20T00:03:49Z"}],"graph_snapshots":[{"event_id":"sha256:01f8628caee42b8fccd9d035e0afc9170bdc3b15befccbfe62e0e3aa0b82502b","target":"graph","created_at":"2026-05-20T00:03:49Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"no evaluated model clears 80%, with Claude-Opus-4.6 leading at 77.8%, revealing that current frontier models still fail to comply with observation contracts"},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The 33 dual-axis tasks and their failure labels drawn from real-world API specifications sufficiently capture the observation-contract compliance problem that arises in deployed tool-augmented agents."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ContractBench shows that LLM agents frequently violate observation contracts by using expired artifacts or corrupting their byte integrity, with no model exceeding 80% success and notable scaling irregularities across families."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLM agents must preserve observation contracts like tokens and presigned URLs, yet current models routinely fail at this separate capability."}],"snapshot_sha256":"8a260291f250f1411f401f72766fe4b8cd014846e2636cda4027f988ebe50706"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"4484454e7eed5f4c186d02123f11525561d4f7836ef5bff1e3f2dabc305f3017"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T23:01:55.266047Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T23:01:19.691587Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T22:01:57.826415Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T21:33:23.771987Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.17281/integrity.json","findings":[],"snapshot_sha256":"7aa72b3cde359d45cbd2ac55640b44b9d8777d9c8f58c8e1c83134c0c50ccae9","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Tool-augmented LLM agents call APIs whose intermediate outputs, such as presigned URLs, session tokens, and OAuth state parameters, are observation contracts: artifacts whose later use is constrained by the external system that produced them. We show that observation contract compliance (preserving the temporal validity and byte-level integrity) is an emergent, regression-prone capability: it is neither guaranteed by general tool-use ability nor consistently improved by larger or newer models. To measure this, we introduce ContractBench, a benchmark of 33 dual-axis tasks that probe two orthogo","authors_text":"Arkaprava De, Hanwen Xing, Hao Chen, Jicheng Wang, Yifeng He, Zili Wang","cross_cats":["cs.AI"],"headline":"LLM agents must preserve observation contracts like tokens and presigned URLs, yet current models routinely fail at this separate capability.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-17T06:37:04Z","title":"ContractBench: Can LLM Agents Preserve Observation Contracts?"},"references":{"count":44,"internal_anchors":1,"resolved_work":44,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"2026 , eprint =","work_id":"f42c6a1e-2720-44a2-9b07-d0d85070e8f4","year":2026},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"2026 , eprint =","work_id":"4de62ded-8c5f-4635-892f-e124acbaf237","year":2026},{"cited_arxiv_id":"","doi":"10.1145/365230.365252","is_internal_anchor":false,"ref_index":3,"title":"Programming semantics for multiprogrammed computations.Commun","work_id":"853d34ab-9346-40c5-aa7c-a65cb42958e5","year":1966},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"2024 , url =","work_id":"e979d9bc-7642-42e6-9755-75dd0e116f94","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"AgentBench: Evaluating","work_id":"98c5d123-b91d-45c7-ae4f-14f91745e06c","year":2024}],"snapshot_sha256":"3e322bba5d1c93429439a374bdd7e24c3c321ac4bb6abcca52e419396e03bdf0"},"source":{"id":"2605.17281","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T22:56:39.462455Z","id":"5bc5e912-8b8f-4b3c-8c96-72ebaa0fab0b","model_set":{"reader":"grok-4.3"},"one_line_summary":"ContractBench shows that LLM agents frequently violate observation contracts by using expired artifacts or corrupting their byte integrity, with no model exceeding 80% success and notable scaling irregularities across families.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"LLM agents must preserve observation contracts like tokens and presigned URLs, yet current models routinely fail at this separate capability.","strongest_claim":"no evaluated model clears 80%, with Claude-Opus-4.6 leading at 77.8%, revealing that current frontier models still fail to comply with observation contracts","weakest_assumption":"The 33 dual-axis tasks and their failure labels drawn from real-world API specifications sufficiently capture the observation-contract compliance problem that arises in deployed tool-augmented agents."}},"verdict_id":"5bc5e912-8b8f-4b3c-8c96-72ebaa0fab0b"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:de9ea484f5e4ce3c5ade87c74c4b167b766d6a9bffb5c65e09267d2883da657e","target":"record","created_at":"2026-05-20T00:03:49Z","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":"a41d80bf262fe501d30e568d92f73f98eb919970804ba764aba2c6acf257fb86","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-05-17T06:37:04Z","title_canon_sha256":"29507e0cd5028ab233f6d828aa1c66393e13166fb034674024c2886cf2f0971d"},"schema_version":"1.0","source":{"id":"2605.17281","kind":"arxiv","version":1}},"canonical_sha256":"4612a3b2ff4ff793756ee31807bdbeef45529d1249e54c0aff62f318755c7218","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4612a3b2ff4ff793756ee31807bdbeef45529d1249e54c0aff62f318755c7218","first_computed_at":"2026-05-20T00:03:49.580728Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:49.580728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/+pyzxEfc4HjbakNULxspDd08eAlkmpylY1DDMON5Ih0rSewBYOmokjhMntlWphJMvbzknk4qf05aszY4v9dCA==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:49.581546Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.17281","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:de9ea484f5e4ce3c5ade87c74c4b167b766d6a9bffb5c65e09267d2883da657e","sha256:01f8628caee42b8fccd9d035e0afc9170bdc3b15befccbfe62e0e3aa0b82502b"],"state_sha256":"fc30cdf870a80deedffe11d8886fc7d7dce9ddbd3d4bc648903f4806d7ea39e3"}