{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YIB3VNHIZM7QCNUYVWDGPSXWNY","short_pith_number":"pith:YIB3VNHI","schema_version":"1.0","canonical_sha256":"c203bab4e8cb3f013698ad8667caf66e1b73fbcfd739dd8232b833095f53df3d","source":{"kind":"arxiv","id":"2505.17107","version":1},"attestation_state":"computed","paper":{"title":"CRAKEN: Cybersecurity LLM Agent with Knowledge-Based Execution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MA"],"primary_cat":"cs.CR","authors_text":"Brendan Dolan-Gavitt, Farshad Khorrami, Haoran Xi, Kimberly Milner, Meet Udeshi, Minghao Shao, Muhammad Shafique, Nanda Rani, Prashanth Krishnamurthy, Ramesh Karri, Sandeep Kumar Shukla, Venkata Sai Charan Putrevu","submitted_at":"2025-05-21T11:01:11Z","abstract_excerpt":"Large Language Model (LLM) agents can automate cybersecurity tasks and can adapt to the evolving cybersecurity landscape without re-engineering. While LLM agents have demonstrated cybersecurity capabilities on Capture-The-Flag (CTF) competitions, they have two key limitations: accessing latest cybersecurity expertise beyond training data, and integrating new knowledge into complex task planning. Knowledge-based approaches that incorporate technical understanding into the task-solving automation can tackle these limitations. We present CRAKEN, a knowledge-based LLM agent framework that improves"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2505.17107","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-05-21T11:01:11Z","cross_cats_sorted":["cs.AI","cs.LG","cs.MA"],"title_canon_sha256":"fa5c1ab5772b12064b5faf284c209b8d43abe05796fa709be7fbe5e6a71b2164","abstract_canon_sha256":"40747b4978c30a6186299b680c5c7077ebd9941424e95b71d0aff74139f54214"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:07:47.186726Z","signature_b64":"RHWRTQvp701GJv2V3RpTTs4DT10ssksDjt8VeduN/e/X1HOZ0euISDfnnV0rUMjU2hwKfgMZoC+rLZCVcsOfBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c203bab4e8cb3f013698ad8667caf66e1b73fbcfd739dd8232b833095f53df3d","last_reissued_at":"2026-07-05T11:07:47.186226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:07:47.186226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CRAKEN: Cybersecurity LLM Agent with Knowledge-Based Execution","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.MA"],"primary_cat":"cs.CR","authors_text":"Brendan Dolan-Gavitt, Farshad Khorrami, Haoran Xi, Kimberly Milner, Meet Udeshi, Minghao Shao, Muhammad Shafique, Nanda Rani, Prashanth Krishnamurthy, Ramesh Karri, Sandeep Kumar Shukla, Venkata Sai Charan Putrevu","submitted_at":"2025-05-21T11:01:11Z","abstract_excerpt":"Large Language Model (LLM) agents can automate cybersecurity tasks and can adapt to the evolving cybersecurity landscape without re-engineering. While LLM agents have demonstrated cybersecurity capabilities on Capture-The-Flag (CTF) competitions, they have two key limitations: accessing latest cybersecurity expertise beyond training data, and integrating new knowledge into complex task planning. Knowledge-based approaches that incorporate technical understanding into the task-solving automation can tackle these limitations. We present CRAKEN, a knowledge-based LLM agent framework that improves"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.17107","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/2505.17107/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2505.17107","created_at":"2026-07-05T11:07:47.186288+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.17107v1","created_at":"2026-07-05T11:07:47.186288+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.17107","created_at":"2026-07-05T11:07:47.186288+00:00"},{"alias_kind":"pith_short_12","alias_value":"YIB3VNHIZM7Q","created_at":"2026-07-05T11:07:47.186288+00:00"},{"alias_kind":"pith_short_16","alias_value":"YIB3VNHIZM7QCNUY","created_at":"2026-07-05T11:07:47.186288+00:00"},{"alias_kind":"pith_short_8","alias_value":"YIB3VNHI","created_at":"2026-07-05T11:07:47.186288+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":11,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25332","citing_title":"Decoupling Reconnaissance and Exploitation: Measuring the Capability Boundaries of LLM-Based Web Penetration Testing","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26195","citing_title":"CyberEvolver: Structured Self-Evolution for Cybersecurity Agents On the Fly","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29269","citing_title":"HunterAgent: Neuro-Symbolic Attack Trace Reconstruction under Anti-Forensics","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29963","citing_title":"Honeyval: A Comprehensive Evaluation Framework for LLM-powered HTTP Honeypots","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15503","citing_title":"uGen: An Agentic Framework for Generating Microarchitectural Attack PoCs","ref_index":43,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15097","citing_title":"Veritas: Grounding LLM Agents for Reliable Vulnerability Reasoning over Stripped Binaries","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2605.11504","citing_title":"CTFusion: A CTF-based Benchmark for LLM Agent Evaluation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10597","citing_title":"CrackMeBench: Binary Reverse Engineering for Agents","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.24184","citing_title":"Dynamic Cyber Ranges","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17948","citing_title":"RAVEN: Retrieval-Augmented Vulnerability Exploration Network for Memory Corruption Analysis in User Code and Binary Programs","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17159","citing_title":"Systematic Capability Benchmarking of Frontier Large Language Models for Offensive Cyber Tasks","ref_index":8,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY","json":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY.json","graph_json":"https://pith.science/api/pith-number/YIB3VNHIZM7QCNUYVWDGPSXWNY/graph.json","events_json":"https://pith.science/api/pith-number/YIB3VNHIZM7QCNUYVWDGPSXWNY/events.json","paper":"https://pith.science/paper/YIB3VNHI"},"agent_actions":{"view_html":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY","download_json":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY.json","view_paper":"https://pith.science/paper/YIB3VNHI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.17107&json=true","fetch_graph":"https://pith.science/api/pith-number/YIB3VNHIZM7QCNUYVWDGPSXWNY/graph.json","fetch_events":"https://pith.science/api/pith-number/YIB3VNHIZM7QCNUYVWDGPSXWNY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY/action/storage_attestation","attest_author":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY/action/author_attestation","sign_citation":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY/action/citation_signature","submit_replication":"https://pith.science/pith/YIB3VNHIZM7QCNUYVWDGPSXWNY/action/replication_record"}},"created_at":"2026-07-05T11:07:47.186288+00:00","updated_at":"2026-07-05T11:07:47.186288+00:00"}