{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:ZSZB5PARFDOFWWLE54TLHT4432","short_pith_number":"pith:ZSZB5PAR","canonical_record":{"source":{"id":"2607.22368","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-24T14:55:19Z","cross_cats_sorted":[],"title_canon_sha256":"744ca0099228b3a94fe680cc78658da5542d00d2c3b22c86cf7869e32010f917","abstract_canon_sha256":"192b91dec1d46a7aa7df6c5b508523b56284a6091cbfe5c86ce24507657aeaed"},"schema_version":"1.0"},"canonical_sha256":"ccb21ebc1128dc5b5964ef26b3cf9cdead83f7312955a901fe1c18764c50bf97","source":{"kind":"arxiv","id":"2607.22368","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.22368","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"arxiv_version","alias_value":"2607.22368v1","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22368","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"pith_short_12","alias_value":"ZSZB5PARFDOF","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"pith_short_16","alias_value":"ZSZB5PARFDOFWWLE","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"pith_short_8","alias_value":"ZSZB5PAR","created_at":"2026-07-27T01:21:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:ZSZB5PARFDOFWWLE54TLHT4432","target":"record","payload":{"canonical_record":{"source":{"id":"2607.22368","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-24T14:55:19Z","cross_cats_sorted":[],"title_canon_sha256":"744ca0099228b3a94fe680cc78658da5542d00d2c3b22c86cf7869e32010f917","abstract_canon_sha256":"192b91dec1d46a7aa7df6c5b508523b56284a6091cbfe5c86ce24507657aeaed"},"schema_version":"1.0"},"canonical_sha256":"ccb21ebc1128dc5b5964ef26b3cf9cdead83f7312955a901fe1c18764c50bf97","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-27T01:21:16.918660Z","signature_b64":"ipFF61sHT3IuvFHC+LyUyAojy+oMc/gJd7k2wJaUFktml7+8w86Efakrb5MTTUZRoQyRqtVbIY+W7WeOIfgdAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ccb21ebc1128dc5b5964ef26b3cf9cdead83f7312955a901fe1c18764c50bf97","last_reissued_at":"2026-07-27T01:21:16.917794Z","signature_status":"signed_v1","first_computed_at":"2026-07-27T01:21:16.917794Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.22368","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":"2026-07-27T01:21:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h3pwnV+IEZljFRvovvTqd8ZY8gOR0IQrhUY1TSL7kGE8sHRZzm7OYwYfFwh+puz9YbMfxpJpjymAKAws1htZCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T14:52:01.355216Z"},"content_sha256":"868f6ea9ee18da98a36fe779ade6e18ff773f19bb6458536242902dc763a6b1c","schema_version":"1.0","event_id":"sha256:868f6ea9ee18da98a36fe779ade6e18ff773f19bb6458536242902dc763a6b1c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:ZSZB5PARFDOFWWLE54TLHT4432","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Do Agent Benchmarks Measure Capability? Protocol Validity in the Age of Agentic AI","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bing Luo, Hanck Chen, Jiaqi Shao, Maxm Pan, Wei Zhang","submitted_at":"2026-07-24T14:55:19Z","abstract_excerpt":"Agent benchmarks increasingly evaluate repository editing, web research, terminal use, and long-horizon interaction. Their scores support capability claims only when the evaluation protocol keeps the intended capability necessary for success. Recent reward-hacking benchmarks and system reports show that agents can instead recover public solutions, read evaluation artifacts, infer generator structure, manipulate feedback, or benefit from invalid scoring paths; existing responses do not provide a common procedure for attributing these shortcuts and quantifying their effect across benchmarks. We "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22368","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/2607.22368/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":"2026-07-27T01:21:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7hDlnps3HZQ3mdFLX9p0dyJhJOMypdzRvSMbio1G0X0K77/YCEBLEaB1Wh4tRWsplVfYNFjez4Pkj0fDmkwmAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T14:52:01.355798Z"},"content_sha256":"6e73a3e3714f23df4f76a8f16b1682d610feb35c83228cb4faa1a9eb1aae3a3b","schema_version":"1.0","event_id":"sha256:6e73a3e3714f23df4f76a8f16b1682d610feb35c83228cb4faa1a9eb1aae3a3b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZSZB5PARFDOFWWLE54TLHT4432/bundle.json","state_url":"https://pith.science/pith/ZSZB5PARFDOFWWLE54TLHT4432/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZSZB5PARFDOFWWLE54TLHT4432/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":"2026-08-01T14:52:01Z","links":{"resolver":"https://pith.science/pith/ZSZB5PARFDOFWWLE54TLHT4432","bundle":"https://pith.science/pith/ZSZB5PARFDOFWWLE54TLHT4432/bundle.json","state":"https://pith.science/pith/ZSZB5PARFDOFWWLE54TLHT4432/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZSZB5PARFDOFWWLE54TLHT4432/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:ZSZB5PARFDOFWWLE54TLHT4432","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":"192b91dec1d46a7aa7df6c5b508523b56284a6091cbfe5c86ce24507657aeaed","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-24T14:55:19Z","title_canon_sha256":"744ca0099228b3a94fe680cc78658da5542d00d2c3b22c86cf7869e32010f917"},"schema_version":"1.0","source":{"id":"2607.22368","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.22368","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"arxiv_version","alias_value":"2607.22368v1","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.22368","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"pith_short_12","alias_value":"ZSZB5PARFDOF","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"pith_short_16","alias_value":"ZSZB5PARFDOFWWLE","created_at":"2026-07-27T01:21:16Z"},{"alias_kind":"pith_short_8","alias_value":"ZSZB5PAR","created_at":"2026-07-27T01:21:16Z"}],"graph_snapshots":[{"event_id":"sha256:6e73a3e3714f23df4f76a8f16b1682d610feb35c83228cb4faa1a9eb1aae3a3b","target":"graph","created_at":"2026-07-27T01:21:16Z","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/2607.22368/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Agent benchmarks increasingly evaluate repository editing, web research, terminal use, and long-horizon interaction. Their scores support capability claims only when the evaluation protocol keeps the intended capability necessary for success. Recent reward-hacking benchmarks and system reports show that agents can instead recover public solutions, read evaluation artifacts, infer generator structure, manipulate feedback, or benefit from invalid scoring paths; existing responses do not provide a common procedure for attributing these shortcuts and quantifying their effect across benchmarks. We ","authors_text":"Bing Luo, Hanck Chen, Jiaqi Shao, Maxm Pan, Wei Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-24T14:55:19Z","title":"Do Agent Benchmarks Measure Capability? Protocol Validity in the Age of Agentic AI"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.22368","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":"sha256:868f6ea9ee18da98a36fe779ade6e18ff773f19bb6458536242902dc763a6b1c","target":"record","created_at":"2026-07-27T01:21:16Z","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":"192b91dec1d46a7aa7df6c5b508523b56284a6091cbfe5c86ce24507657aeaed","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-24T14:55:19Z","title_canon_sha256":"744ca0099228b3a94fe680cc78658da5542d00d2c3b22c86cf7869e32010f917"},"schema_version":"1.0","source":{"id":"2607.22368","kind":"arxiv","version":1}},"canonical_sha256":"ccb21ebc1128dc5b5964ef26b3cf9cdead83f7312955a901fe1c18764c50bf97","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ccb21ebc1128dc5b5964ef26b3cf9cdead83f7312955a901fe1c18764c50bf97","first_computed_at":"2026-07-27T01:21:16.917794Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-27T01:21:16.917794Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ipFF61sHT3IuvFHC+LyUyAojy+oMc/gJd7k2wJaUFktml7+8w86Efakrb5MTTUZRoQyRqtVbIY+W7WeOIfgdAQ==","signature_status":"signed_v1","signed_at":"2026-07-27T01:21:16.918660Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.22368","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:868f6ea9ee18da98a36fe779ade6e18ff773f19bb6458536242902dc763a6b1c","sha256:6e73a3e3714f23df4f76a8f16b1682d610feb35c83228cb4faa1a9eb1aae3a3b"],"state_sha256":"4d100c9bd0feacb3ecf0ab00543b83d29f76bb02e2844e931f148a3b64d34301"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"97hSVJPCLD12aQEfRw67D47eqR0mXKJXlbtrcNm7njUqr0vuyGqHbNUuoPmF7qPxeVxfR+LO+9HNWf5YjaGMAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T14:52:01.359441Z","bundle_sha256":"09d047a98ef2869bb37e1f1cde4b9723d9338ec82678dd53dbc0ea84c91d8bcf"}}