{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:67TQULDDWOXFJFGIMJC26IV5IQ","short_pith_number":"pith:67TQULDD","schema_version":"1.0","canonical_sha256":"f7e70a2c63b3ae5494c86245af22bd442407d8ab7fa3a0e07366c142f8405120","source":{"kind":"arxiv","id":"2608.13547","version":1},"attestation_state":"computed","paper":{"title":"QuoteBench: How Matched Scores Can Hide Command-Path Failures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.AI","authors_text":"Shangao Li, Volker Tresp, Yao Zhang, Yuanyuan Yang","submitted_at":"2026-08-13T17:57:20Z","abstract_excerpt":"LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched execution scores alone cannot distinguish command-generation errors from failures introduced after generation. QuoteBench measures this boundary with exact final-state validation on 56 one-shot tasks from 14 incident-derived families, crossing the generation contract with the execution transport around one deliberately unescaped added parser. Escaping at the interpolation point reproduces each replayed reply's raw-path outcome, so any recovery under a disclosed boundary must com"},"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":"2608.13547","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-13T17:57:20Z","cross_cats_sorted":["cs.SE"],"title_canon_sha256":"957279685f20f5f37c455a9448de56f1bf962f446825c2f07ae8a369acd2e38a","abstract_canon_sha256":"97a02a5cfcd84a676a1bebb7b52ca11d729ff9484ef0b0e518ead0e1d7f775cc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-14T01:25:02.372183Z","signature_b64":"xrLZUIjTlDvOrt5a8OgkkabN624J0IRIBLOCKb3nzHrxvQnIpfG3VP4D+x0IBs+jtBvlAiEFdHoOzwFvD7BMCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f7e70a2c63b3ae5494c86245af22bd442407d8ab7fa3a0e07366c142f8405120","last_reissued_at":"2026-08-14T01:25:02.369728Z","signature_status":"signed_v1","first_computed_at":"2026-08-14T01:25:02.369728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"QuoteBench: How Matched Scores Can Hide Command-Path Failures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.SE"],"primary_cat":"cs.AI","authors_text":"Shangao Li, Volker Tresp, Yao Zhang, Yuanyuan Yang","submitted_at":"2026-08-13T17:57:20Z","abstract_excerpt":"LLM coding agents issue Bash commands through interfaces that may serialize, wrap, and reparse model output. Matched execution scores alone cannot distinguish command-generation errors from failures introduced after generation. QuoteBench measures this boundary with exact final-state validation on 56 one-shot tasks from 14 incident-derived families, crossing the generation contract with the execution transport around one deliberately unescaped added parser. Escaping at the interpolation point reproduces each replayed reply's raw-path outcome, so any recovery under a disclosed boundary must com"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13547","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/2608.13547/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":"2608.13547","created_at":"2026-08-14T01:25:02.371438+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.13547v1","created_at":"2026-08-14T01:25:02.371438+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13547","created_at":"2026-08-14T01:25:02.371438+00:00"},{"alias_kind":"pith_short_12","alias_value":"67TQULDDWOXF","created_at":"2026-08-14T01:25:02.371438+00:00"},{"alias_kind":"pith_short_16","alias_value":"67TQULDDWOXFJFGI","created_at":"2026-08-14T01:25:02.371438+00:00"},{"alias_kind":"pith_short_8","alias_value":"67TQULDD","created_at":"2026-08-14T01:25:02.371438+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/67TQULDDWOXFJFGIMJC26IV5IQ","json":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ.json","graph_json":"https://pith.science/api/pith-number/67TQULDDWOXFJFGIMJC26IV5IQ/graph.json","events_json":"https://pith.science/api/pith-number/67TQULDDWOXFJFGIMJC26IV5IQ/events.json","paper":"https://pith.science/paper/67TQULDD"},"agent_actions":{"view_html":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ","download_json":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ.json","view_paper":"https://pith.science/paper/67TQULDD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.13547&json=true","fetch_graph":"https://pith.science/api/pith-number/67TQULDDWOXFJFGIMJC26IV5IQ/graph.json","fetch_events":"https://pith.science/api/pith-number/67TQULDDWOXFJFGIMJC26IV5IQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/action/storage_attestation","attest_author":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/action/author_attestation","sign_citation":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/action/citation_signature","submit_replication":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/action/replication_record"}},"created_at":"2026-08-14T01:25:02.371438+00:00","updated_at":"2026-08-14T01:25:02.371438+00:00"}