{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:67TQULDDWOXFJFGIMJC26IV5IQ","short_pith_number":"pith:67TQULDD","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"},"canonical_sha256":"f7e70a2c63b3ae5494c86245af22bd442407d8ab7fa3a0e07366c142f8405120","source":{"kind":"arxiv","id":"2608.13547","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.13547","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"arxiv_version","alias_value":"2608.13547v1","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13547","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"pith_short_12","alias_value":"67TQULDDWOXF","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"pith_short_16","alias_value":"67TQULDDWOXFJFGI","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"pith_short_8","alias_value":"67TQULDD","created_at":"2026-08-14T01:25:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:67TQULDDWOXFJFGIMJC26IV5IQ","target":"record","payload":{"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"},"canonical_sha256":"f7e70a2c63b3ae5494c86245af22bd442407d8ab7fa3a0e07366c142f8405120","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"},"source_kind":"arxiv","source_id":"2608.13547","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-14T01:25:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9VFaxlsq9UHFFbC0de0TuqnTmF69kWSDbLWN4kcXTBfJlyL4JKuZxOoqr/6ImmDdacNGWbzPecMmmsMhIiEUDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T05:46:57.061127Z"},"content_sha256":"f025833b23eb2d4e60a38cc40dccf900506e7763cbcb28b016036e63f8f7a4d2","schema_version":"1.0","event_id":"sha256:f025833b23eb2d4e60a38cc40dccf900506e7763cbcb28b016036e63f8f7a4d2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:67TQULDDWOXFJFGIMJC26IV5IQ","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-14T01:25:02Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TZKEC83gnxznHcPO9PLiarOT+12/5/iX01DAV1+LcJTxL8b8NztOZSZhSsAjAjlY54EX+kWJMayuPi2QYb6oAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-15T05:46:57.061649Z"},"content_sha256":"1765eb628363ac5720b99ee27680b66d7b12c5eb383059a699e1047c03233579","schema_version":"1.0","event_id":"sha256:1765eb628363ac5720b99ee27680b66d7b12c5eb383059a699e1047c03233579"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/bundle.json","state_url":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/67TQULDDWOXFJFGIMJC26IV5IQ/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":"2026-08-15T05:46:57Z","links":{"resolver":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ","bundle":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/bundle.json","state":"https://pith.science/pith/67TQULDDWOXFJFGIMJC26IV5IQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/67TQULDDWOXFJFGIMJC26IV5IQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:67TQULDDWOXFJFGIMJC26IV5IQ","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":"97a02a5cfcd84a676a1bebb7b52ca11d729ff9484ef0b0e518ead0e1d7f775cc","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-13T17:57:20Z","title_canon_sha256":"957279685f20f5f37c455a9448de56f1bf962f446825c2f07ae8a369acd2e38a"},"schema_version":"1.0","source":{"id":"2608.13547","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.13547","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"arxiv_version","alias_value":"2608.13547v1","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.13547","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"pith_short_12","alias_value":"67TQULDDWOXF","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"pith_short_16","alias_value":"67TQULDDWOXFJFGI","created_at":"2026-08-14T01:25:02Z"},{"alias_kind":"pith_short_8","alias_value":"67TQULDD","created_at":"2026-08-14T01:25:02Z"}],"graph_snapshots":[{"event_id":"sha256:1765eb628363ac5720b99ee27680b66d7b12c5eb383059a699e1047c03233579","target":"graph","created_at":"2026-08-14T01:25:02Z","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/2608.13547/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Shangao Li, Volker Tresp, Yao Zhang, Yuanyuan Yang","cross_cats":["cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-13T17:57:20Z","title":"QuoteBench: How Matched Scores Can Hide Command-Path Failures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.13547","kind":"arxiv","version":1},"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:f025833b23eb2d4e60a38cc40dccf900506e7763cbcb28b016036e63f8f7a4d2","target":"record","created_at":"2026-08-14T01:25:02Z","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":"97a02a5cfcd84a676a1bebb7b52ca11d729ff9484ef0b0e518ead0e1d7f775cc","cross_cats_sorted":["cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-13T17:57:20Z","title_canon_sha256":"957279685f20f5f37c455a9448de56f1bf962f446825c2f07ae8a369acd2e38a"},"schema_version":"1.0","source":{"id":"2608.13547","kind":"arxiv","version":1}},"canonical_sha256":"f7e70a2c63b3ae5494c86245af22bd442407d8ab7fa3a0e07366c142f8405120","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7e70a2c63b3ae5494c86245af22bd442407d8ab7fa3a0e07366c142f8405120","first_computed_at":"2026-08-14T01:25:02.369728Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-14T01:25:02.369728Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"xrLZUIjTlDvOrt5a8OgkkabN624J0IRIBLOCKb3nzHrxvQnIpfG3VP4D+x0IBs+jtBvlAiEFdHoOzwFvD7BMCg==","signature_status":"signed_v1","signed_at":"2026-08-14T01:25:02.372183Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.13547","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f025833b23eb2d4e60a38cc40dccf900506e7763cbcb28b016036e63f8f7a4d2","sha256:1765eb628363ac5720b99ee27680b66d7b12c5eb383059a699e1047c03233579"],"state_sha256":"e344fa7e30b06469ead0c853f7b1b341e1c47eb824fcba46644b0abb870dac90"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k6S7A0oLm8EmXCtnEbLnXMF9jICi0nVKWoQsCgk/bL5YnYlxMWzq+8f6VDrWH2NtQw2Q2SOymWEoTJLlIdu1AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-15T05:46:57.066351Z","bundle_sha256":"04af7d6c83b8ddb79902d14a872451d3f9209733537dc744c968e434ed63e6e8"}}