{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:VVBYJTCUTSWUX3LUCOBIJ2VPFL","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":"843db3d05659d325314ad6498d61b32666720d371ee873208f05e50d8f5d8d4c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2024-12-02T18:28:18Z","title_canon_sha256":"064c861f4f5f6ee9ebc47165389a0122ae81c4386f28fcab158fc45ade7762e4"},"schema_version":"1.0","source":{"id":"2412.01778","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.01778","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"arxiv_version","alias_value":"2412.01778v1","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01778","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"pith_short_12","alias_value":"VVBYJTCUTSWU","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"pith_short_16","alias_value":"VVBYJTCUTSWUX3LU","created_at":"2026-07-05T09:43:19Z"},{"alias_kind":"pith_short_8","alias_value":"VVBYJTCU","created_at":"2026-07-05T09:43:19Z"}],"graph_snapshots":[{"event_id":"sha256:2ec3c5f32f684462625143fd21d37a9a6ad3a0000396b1f04d06525441953364","target":"graph","created_at":"2026-07-05T09:43:19Z","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/2412.01778/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce HackSynth, a novel Large Language Model (LLM)-based agent capable of autonomous penetration testing. HackSynth's dual-module architecture includes a Planner and a Summarizer, which enable it to generate commands and process feedback iteratively. To benchmark HackSynth, we propose two new Capture The Flag (CTF)-based benchmark sets utilizing the popular platforms PicoCTF and OverTheWire. These benchmarks include two hundred challenges across diverse domains and difficulties, providing a standardized framework for evaluating LLM-based penetration testing agents. Based on these bench","authors_text":"Andr\\'as Luk\\'acs, David Imolai, Lajos Muzsai","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2024-12-02T18:28:18Z","title":"HackSynth: LLM Agent and Evaluation Framework for Autonomous Penetration Testing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01778","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:fa29e213daf43d407349b26651702769c6e50772789e27d0db146140aea7db0d","target":"record","created_at":"2026-07-05T09:43:19Z","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":"843db3d05659d325314ad6498d61b32666720d371ee873208f05e50d8f5d8d4c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2024-12-02T18:28:18Z","title_canon_sha256":"064c861f4f5f6ee9ebc47165389a0122ae81c4386f28fcab158fc45ade7762e4"},"schema_version":"1.0","source":{"id":"2412.01778","kind":"arxiv","version":1}},"canonical_sha256":"ad4384cc549cad4bed74138284eaaf2ac7c4d790b0ea453da4d5e819c74f8d8c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ad4384cc549cad4bed74138284eaaf2ac7c4d790b0ea453da4d5e819c74f8d8c","first_computed_at":"2026-07-05T09:43:19.472075Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:43:19.472075Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"UN8cNpITFez2TbYEPtJqyHC8FzzhYcIE2FCtPnUnisJieRXZxsydKCT+RIKDJHJV+Bi9vOq+F+CJQYDqRE+qDw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:43:19.472608Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.01778","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fa29e213daf43d407349b26651702769c6e50772789e27d0db146140aea7db0d","sha256:2ec3c5f32f684462625143fd21d37a9a6ad3a0000396b1f04d06525441953364"],"state_sha256":"0decfc7385a174d4d9bd27c8f06656d868e3af85b1f9a15da9875a42ff0c5aba"}