{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:VOT4SMH3RKE2VGWPQTWCOQME7T","short_pith_number":"pith:VOT4SMH3","canonical_record":{"source":{"id":"2605.21497","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-04-29T09:42:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e384cb8310393c6ffb75808481a7759f88af0b127fb75c37391edf9a5fb40504","abstract_canon_sha256":"258f95297bc3adf1f540f6d76322e12f180a7783034f9f52261d09c12e027535"},"schema_version":"1.0"},"canonical_sha256":"aba7c930fb8a89aa9acf84ec274184fcc4b36aa87e81580e86db2db5701705b8","source":{"kind":"arxiv","id":"2605.21497","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21497","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21497v1","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21497","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"pith_short_12","alias_value":"VOT4SMH3RKE2","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"pith_short_16","alias_value":"VOT4SMH3RKE2VGWP","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"pith_short_8","alias_value":"VOT4SMH3","created_at":"2026-05-22T00:02:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:VOT4SMH3RKE2VGWPQTWCOQME7T","target":"record","payload":{"canonical_record":{"source":{"id":"2605.21497","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-04-29T09:42:36Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e384cb8310393c6ffb75808481a7759f88af0b127fb75c37391edf9a5fb40504","abstract_canon_sha256":"258f95297bc3adf1f540f6d76322e12f180a7783034f9f52261d09c12e027535"},"schema_version":"1.0"},"canonical_sha256":"aba7c930fb8a89aa9acf84ec274184fcc4b36aa87e81580e86db2db5701705b8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T00:02:26.986898Z","signature_b64":"GPAIDjKXDQcfQ3c9TICmboLCDytosMbtMk2jRDdinIrX4knAWSqR+2oCd9qOHObrQVFKwi8fh8V/zdVQbYZODw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aba7c930fb8a89aa9acf84ec274184fcc4b36aa87e81580e86db2db5701705b8","last_reissued_at":"2026-05-22T00:02:26.986435Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T00:02:26.986435Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.21497","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":"2026-05-22T00:02:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w6VMDyezUdFMqj7FLDZCpz6WZb4118KGukzKMsSiGjCOg1KXeT8RMgjewyD35Ht2spuok7fIyRD5JLDCzdfiAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T09:13:13.164186Z"},"content_sha256":"ead1d6ce816e1239c22a84e8c6e233080f58d9c5dc77db1575660a61d4dca86a","schema_version":"1.0","event_id":"sha256:ead1d6ce816e1239c22a84e8c6e233080f58d9c5dc77db1575660a61d4dca86a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:VOT4SMH3RKE2VGWPQTWCOQME7T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Autonomous LLM Agents & CTFs: A Second Look","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Dario Rossi, Idilio Drago, Marco Mellia, Matteo Boffa, Thanh Minh Bui, Youness Bouchari","submitted_at":"2026-04-29T09:42:36Z","abstract_excerpt":"Large Language Model (LLM) agents are increasingly proposed to automate offensive security tasks, with recent studies reporting near human-level success rates in Capture-the-Flag (CTF) challenges. We here revisit these results, providing a second look at these claims. We engineer different agent architectures of increasing complexity and modularity on 30 web-based CTFs challenges spanning 14 vulnerability classes. We instantiate these agents with multiple LLM backbones, and compare them with claude-code, a general-purpose agent that automatically determines its internal architecture. Our evalu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21497","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/2605.21497/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":"2026-05-22T00:02:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yHmC8pUPI1VlU2y6Q+9/eymXXDZQoPojGw+0MQqw1gYkmsotkwDwdrEmNFvkA9bgu8puLGLwas7LmTjp3T5FBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T09:13:13.164986Z"},"content_sha256":"79de55bd1ae5f27d48a9c2373a0892843fe9e61c735c2b353109b5b7852f0766","schema_version":"1.0","event_id":"sha256:79de55bd1ae5f27d48a9c2373a0892843fe9e61c735c2b353109b5b7852f0766"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VOT4SMH3RKE2VGWPQTWCOQME7T/bundle.json","state_url":"https://pith.science/pith/VOT4SMH3RKE2VGWPQTWCOQME7T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VOT4SMH3RKE2VGWPQTWCOQME7T/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":"2026-05-26T09:13:13Z","links":{"resolver":"https://pith.science/pith/VOT4SMH3RKE2VGWPQTWCOQME7T","bundle":"https://pith.science/pith/VOT4SMH3RKE2VGWPQTWCOQME7T/bundle.json","state":"https://pith.science/pith/VOT4SMH3RKE2VGWPQTWCOQME7T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VOT4SMH3RKE2VGWPQTWCOQME7T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VOT4SMH3RKE2VGWPQTWCOQME7T","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":"258f95297bc3adf1f540f6d76322e12f180a7783034f9f52261d09c12e027535","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-04-29T09:42:36Z","title_canon_sha256":"e384cb8310393c6ffb75808481a7759f88af0b127fb75c37391edf9a5fb40504"},"schema_version":"1.0","source":{"id":"2605.21497","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.21497","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"arxiv_version","alias_value":"2605.21497v1","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.21497","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"pith_short_12","alias_value":"VOT4SMH3RKE2","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"pith_short_16","alias_value":"VOT4SMH3RKE2VGWP","created_at":"2026-05-22T00:02:26Z"},{"alias_kind":"pith_short_8","alias_value":"VOT4SMH3","created_at":"2026-05-22T00:02:26Z"}],"graph_snapshots":[{"event_id":"sha256:79de55bd1ae5f27d48a9c2373a0892843fe9e61c735c2b353109b5b7852f0766","target":"graph","created_at":"2026-05-22T00:02:26Z","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/2605.21497/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Language Model (LLM) agents are increasingly proposed to automate offensive security tasks, with recent studies reporting near human-level success rates in Capture-the-Flag (CTF) challenges. We here revisit these results, providing a second look at these claims. We engineer different agent architectures of increasing complexity and modularity on 30 web-based CTFs challenges spanning 14 vulnerability classes. We instantiate these agents with multiple LLM backbones, and compare them with claude-code, a general-purpose agent that automatically determines its internal architecture. Our evalu","authors_text":"Dario Rossi, Idilio Drago, Marco Mellia, Matteo Boffa, Thanh Minh Bui, Youness Bouchari","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-04-29T09:42:36Z","title":"Autonomous LLM Agents & CTFs: A Second Look"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.21497","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:ead1d6ce816e1239c22a84e8c6e233080f58d9c5dc77db1575660a61d4dca86a","target":"record","created_at":"2026-05-22T00:02:26Z","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":"258f95297bc3adf1f540f6d76322e12f180a7783034f9f52261d09c12e027535","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-04-29T09:42:36Z","title_canon_sha256":"e384cb8310393c6ffb75808481a7759f88af0b127fb75c37391edf9a5fb40504"},"schema_version":"1.0","source":{"id":"2605.21497","kind":"arxiv","version":1}},"canonical_sha256":"aba7c930fb8a89aa9acf84ec274184fcc4b36aa87e81580e86db2db5701705b8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aba7c930fb8a89aa9acf84ec274184fcc4b36aa87e81580e86db2db5701705b8","first_computed_at":"2026-05-22T00:02:26.986435Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T00:02:26.986435Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GPAIDjKXDQcfQ3c9TICmboLCDytosMbtMk2jRDdinIrX4knAWSqR+2oCd9qOHObrQVFKwi8fh8V/zdVQbYZODw==","signature_status":"signed_v1","signed_at":"2026-05-22T00:02:26.986898Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.21497","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ead1d6ce816e1239c22a84e8c6e233080f58d9c5dc77db1575660a61d4dca86a","sha256:79de55bd1ae5f27d48a9c2373a0892843fe9e61c735c2b353109b5b7852f0766"],"state_sha256":"934f9a421483f1acb581c81a2a908ee2cd57245b8b2b396194e99c68a9cc92a8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/6ZyW+sfJcoJ58NhqOnQluGVhgs9lPPu6ZTxQx9iNLUv30+qy84/wvrQldDFUUbzdT4v1TrcI9Fno1Uuzz7NCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T09:13:13.169354Z","bundle_sha256":"58ba6b55ba806b17b38fbb3bfdedd99d9aea3ae0ba118d4623a4bdc7b5256e62"}}