{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:QUHUWX2WAFZUZU2BYL2OBXDLZ3","short_pith_number":"pith:QUHUWX2W","canonical_record":{"source":{"id":"2410.03225","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-10-04T08:24:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d93f47cbbed5c5771d1a82c13d5e8a7cf7e14cbb9a687fd9020c038f015bac47","abstract_canon_sha256":"9bf1298db3bd706ad1ec31e8376c06e623a1d974c2c26e3524326fa50c6e2e87"},"schema_version":"1.0"},"canonical_sha256":"850f4b5f5601734cd341c2f4e0dc6bcef07bf3bd1d1ecff55d96e4ecbc962a17","source":{"kind":"arxiv","id":"2410.03225","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03225","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03225v2","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03225","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"pith_short_12","alias_value":"QUHUWX2WAFZU","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"pith_short_16","alias_value":"QUHUWX2WAFZUZU2B","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"pith_short_8","alias_value":"QUHUWX2W","created_at":"2026-07-05T09:27:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:QUHUWX2WAFZUZU2BYL2OBXDLZ3","target":"record","payload":{"canonical_record":{"source":{"id":"2410.03225","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-10-04T08:24:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"d93f47cbbed5c5771d1a82c13d5e8a7cf7e14cbb9a687fd9020c038f015bac47","abstract_canon_sha256":"9bf1298db3bd706ad1ec31e8376c06e623a1d974c2c26e3524326fa50c6e2e87"},"schema_version":"1.0"},"canonical_sha256":"850f4b5f5601734cd341c2f4e0dc6bcef07bf3bd1d1ecff55d96e4ecbc962a17","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:27:05.425127Z","signature_b64":"ghwlIPEwxtnEDiDpPO6jRF5/d5uTg19URm5tJBfel6yMNmNAnSv+HLXaiq4YVBc6WUv8UIZJckAU3cD8D7weBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"850f4b5f5601734cd341c2f4e0dc6bcef07bf3bd1d1ecff55d96e4ecbc962a17","last_reissued_at":"2026-07-05T09:27:05.424652Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:27:05.424652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.03225","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:27:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JbThVok5ObFuX5K97/7tt+ZF8WOmTdbA0fnQjNR9TJaMTLgMhVkB2OzuJY8LNFiHdyUqSt0ki6s2kdr4P8MBCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T13:25:10.402751Z"},"content_sha256":"6927d9a0ac55c41fe6b05b96d55f7ee768b258d1e16c334b0c639bd224be8af5","schema_version":"1.0","event_id":"sha256:6927d9a0ac55c41fe6b05b96d55f7ee768b258d1e16c334b0c639bd224be8af5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:QUHUWX2WAFZUZU2BYL2OBXDLZ3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AutoPenBench: Benchmarking Generative Agents for Penetration Testing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Alexander Delsanto, Giuseppe Siracusano, Idilio Drago, Luca Gioacchini, Marco Mellia, Roberto Bifulco","submitted_at":"2024-10-04T08:24:15Z","abstract_excerpt":"Generative AI agents, software systems powered by Large Language Models (LLMs), are emerging as a promising approach to automate cybersecurity tasks. Among the others, penetration testing is a challenging field due to the task complexity and the diverse strategies to simulate cyber-attacks. Despite growing interest and initial studies in automating penetration testing with generative agents, there remains a significant gap in the form of a comprehensive and standard framework for their evaluation and development. This paper introduces AutoPenBench, an open benchmark for evaluating generative a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03225","kind":"arxiv","version":2},"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/2410.03225/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":"2026-07-05T09:27:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eDmmt69LPx/G66GnnAc08nIScRUEN4OZXKHQ4aad5W2tCSoZ2Iq8z4DcczbvVexrEi8Py2vqpRweL20PwpX+AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T13:25:10.403262Z"},"content_sha256":"d6d40a43734627af4d40930062ecdcb6f3c8d7b9c5105de897581ce4c109793a","schema_version":"1.0","event_id":"sha256:d6d40a43734627af4d40930062ecdcb6f3c8d7b9c5105de897581ce4c109793a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3/bundle.json","state_url":"https://pith.science/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3/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":"2026-08-09T13:25:10Z","links":{"resolver":"https://pith.science/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3","bundle":"https://pith.science/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3/bundle.json","state":"https://pith.science/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QUHUWX2WAFZUZU2BYL2OBXDLZ3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:QUHUWX2WAFZUZU2BYL2OBXDLZ3","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":"9bf1298db3bd706ad1ec31e8376c06e623a1d974c2c26e3524326fa50c6e2e87","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-10-04T08:24:15Z","title_canon_sha256":"d93f47cbbed5c5771d1a82c13d5e8a7cf7e14cbb9a687fd9020c038f015bac47"},"schema_version":"1.0","source":{"id":"2410.03225","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.03225","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"arxiv_version","alias_value":"2410.03225v2","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.03225","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"pith_short_12","alias_value":"QUHUWX2WAFZU","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"pith_short_16","alias_value":"QUHUWX2WAFZUZU2B","created_at":"2026-07-05T09:27:05Z"},{"alias_kind":"pith_short_8","alias_value":"QUHUWX2W","created_at":"2026-07-05T09:27:05Z"}],"graph_snapshots":[{"event_id":"sha256:d6d40a43734627af4d40930062ecdcb6f3c8d7b9c5105de897581ce4c109793a","target":"graph","created_at":"2026-07-05T09:27:05Z","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/2410.03225/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Generative AI agents, software systems powered by Large Language Models (LLMs), are emerging as a promising approach to automate cybersecurity tasks. Among the others, penetration testing is a challenging field due to the task complexity and the diverse strategies to simulate cyber-attacks. Despite growing interest and initial studies in automating penetration testing with generative agents, there remains a significant gap in the form of a comprehensive and standard framework for their evaluation and development. This paper introduces AutoPenBench, an open benchmark for evaluating generative a","authors_text":"Alexander Delsanto, Giuseppe Siracusano, Idilio Drago, Luca Gioacchini, Marco Mellia, Roberto Bifulco","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-10-04T08:24:15Z","title":"AutoPenBench: Benchmarking Generative Agents for Penetration Testing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.03225","kind":"arxiv","version":2},"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":"sha256:6927d9a0ac55c41fe6b05b96d55f7ee768b258d1e16c334b0c639bd224be8af5","target":"record","created_at":"2026-07-05T09:27:05Z","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":"9bf1298db3bd706ad1ec31e8376c06e623a1d974c2c26e3524326fa50c6e2e87","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2024-10-04T08:24:15Z","title_canon_sha256":"d93f47cbbed5c5771d1a82c13d5e8a7cf7e14cbb9a687fd9020c038f015bac47"},"schema_version":"1.0","source":{"id":"2410.03225","kind":"arxiv","version":2}},"canonical_sha256":"850f4b5f5601734cd341c2f4e0dc6bcef07bf3bd1d1ecff55d96e4ecbc962a17","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"850f4b5f5601734cd341c2f4e0dc6bcef07bf3bd1d1ecff55d96e4ecbc962a17","first_computed_at":"2026-07-05T09:27:05.424652Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:27:05.424652Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ghwlIPEwxtnEDiDpPO6jRF5/d5uTg19URm5tJBfel6yMNmNAnSv+HLXaiq4YVBc6WUv8UIZJckAU3cD8D7weBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:27:05.425127Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.03225","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6927d9a0ac55c41fe6b05b96d55f7ee768b258d1e16c334b0c639bd224be8af5","sha256:d6d40a43734627af4d40930062ecdcb6f3c8d7b9c5105de897581ce4c109793a"],"state_sha256":"87e861b681ae3ef86c51fab01f9d1e8d06d6f4374188d4210f483bdbb66fd26c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ga1EnmKKRZxH4xzbOJ1vQmKs0hb29oInztZ5KX/pQqiAkWKIl76k+IqlhzwqLRv5NRqqY2SKNfEN3Zhj7y84AQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T13:25:10.407481Z","bundle_sha256":"f5fe1f40ff7d2e4183cf788ac23582ab0fd0f564fc2e675ad40ad2b87aa69878"}}