{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:FICEURDQBVNHA7OYL2VX6GZTM4","short_pith_number":"pith:FICEURDQ","schema_version":"1.0","canonical_sha256":"2a044a44700d5a707dd85eab7f1b336738b9dbebd37c8a45fab52d0e12dba2cd","source":{"kind":"arxiv","id":"2607.15684","version":1},"attestation_state":"computed","paper":{"title":"Understanding Agent-Reactive Bugs at the Model-Harness Boundary: An Empirical Study of LLM Agent Issue Reports","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Hengcheng Zhu, Jialun Cao, Jiasi Shen, Jingyi Chen, Shing-Chi Cheung, Songqiang Chen","submitted_at":"2026-07-17T06:53:31Z","abstract_excerpt":"LLM agents span command-line interfaces (e.g., Codex) and agent frameworks (e.g., LangChain), integrating backend LLMs with harness code that parses model outputs, controls agent loops, and manages context. Both the harness and LLM-generated responses jointly shape an agent's execution. This architecture gives rise to bugs that cannot be readily understood by inspecting either component alone, because some bugs occur only when a particular LLM response elicits an abnormal reaction from the agent. Prior empirical studies of agent bugs have largely attributed failures either to limited model cap"},"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":"2607.15684","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2026-07-17T06:53:31Z","cross_cats_sorted":[],"title_canon_sha256":"2febe9416b8e46a688292cf382a6db960566883a58f8b863b125e7e5c469b30e","abstract_canon_sha256":"ce1abba8dc4b03ee382a57deedb028289bbaccabc51840e56f3e837dbcc4615f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-20T01:19:03.988893Z","signature_b64":"I1QFmqd0+W7Nk2qynJio4YDNrZbWJ54t295HPs0ZsFT3bV/V3px8wDBO3Tttkzs0sa9wOs5a2QUkc5f5Is2dAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a044a44700d5a707dd85eab7f1b336738b9dbebd37c8a45fab52d0e12dba2cd","last_reissued_at":"2026-07-20T01:19:03.987951Z","signature_status":"signed_v1","first_computed_at":"2026-07-20T01:19:03.987951Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Understanding Agent-Reactive Bugs at the Model-Harness Boundary: An Empirical Study of LLM Agent Issue Reports","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.SE","authors_text":"Hengcheng Zhu, Jialun Cao, Jiasi Shen, Jingyi Chen, Shing-Chi Cheung, Songqiang Chen","submitted_at":"2026-07-17T06:53:31Z","abstract_excerpt":"LLM agents span command-line interfaces (e.g., Codex) and agent frameworks (e.g., LangChain), integrating backend LLMs with harness code that parses model outputs, controls agent loops, and manages context. Both the harness and LLM-generated responses jointly shape an agent's execution. This architecture gives rise to bugs that cannot be readily understood by inspecting either component alone, because some bugs occur only when a particular LLM response elicits an abnormal reaction from the agent. Prior empirical studies of agent bugs have largely attributed failures either to limited model cap"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.15684","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/2607.15684/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":"2607.15684","created_at":"2026-07-20T01:19:03.988389+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.15684v1","created_at":"2026-07-20T01:19:03.988389+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.15684","created_at":"2026-07-20T01:19:03.988389+00:00"},{"alias_kind":"pith_short_12","alias_value":"FICEURDQBVNH","created_at":"2026-07-20T01:19:03.988389+00:00"},{"alias_kind":"pith_short_16","alias_value":"FICEURDQBVNHA7OY","created_at":"2026-07-20T01:19:03.988389+00:00"},{"alias_kind":"pith_short_8","alias_value":"FICEURDQ","created_at":"2026-07-20T01:19:03.988389+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/FICEURDQBVNHA7OYL2VX6GZTM4","json":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4.json","graph_json":"https://pith.science/api/pith-number/FICEURDQBVNHA7OYL2VX6GZTM4/graph.json","events_json":"https://pith.science/api/pith-number/FICEURDQBVNHA7OYL2VX6GZTM4/events.json","paper":"https://pith.science/paper/FICEURDQ"},"agent_actions":{"view_html":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4","download_json":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4.json","view_paper":"https://pith.science/paper/FICEURDQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.15684&json=true","fetch_graph":"https://pith.science/api/pith-number/FICEURDQBVNHA7OYL2VX6GZTM4/graph.json","fetch_events":"https://pith.science/api/pith-number/FICEURDQBVNHA7OYL2VX6GZTM4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4/action/storage_attestation","attest_author":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4/action/author_attestation","sign_citation":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4/action/citation_signature","submit_replication":"https://pith.science/pith/FICEURDQBVNHA7OYL2VX6GZTM4/action/replication_record"}},"created_at":"2026-07-20T01:19:03.988389+00:00","updated_at":"2026-07-20T01:19:03.988389+00:00"}