{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:5BZ22R6TOXIFOLEGZ6A7J5D5O6","short_pith_number":"pith:5BZ22R6T","schema_version":"1.0","canonical_sha256":"e873ad47d375d0572c86cf81f4f47d77b33f518802ade3c85be0d37b2861d632","source":{"kind":"arxiv","id":"2607.15263","version":1},"attestation_state":"computed","paper":{"title":"Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Blaine Nelson, Paul Kassianik, Yaron Singer","submitted_at":"2026-07-16T17:54:47Z","abstract_excerpt":"Security-agent evaluations commonly measure peak offensive capability under generous inference budgets, emphasizing vulnerability discovery, exploit development, penetration testing, and CTF completion. Such measurements are useful but incomplete: in operational security, every reasoning step, tool call, telemetry query, and enrichment request consumes budget. We evaluate language-model security agents through this cost-success lens on offensive Cybench challenges and defensive Splunk BOTS v1 investigation challenges. Instead of reporting only best-case success, we compare models at fixed cost"},"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":"2607.15263","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2026-07-16T17:54:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"262f09037b27da4de42ad78b0a582198722f204708f1400995554b7a9b330a68","abstract_canon_sha256":"1d99af3f78f6c3f4a395d0e06153603601f8e1f99eb9ea8346d4990fd3f8ba9a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-17T01:22:17.455584Z","signature_b64":"/yUuiTqFfMvuzLo9YEAwBVhPTpOiHhhwWQPkPcpjM+zep4ZIxAGDq/ljVS0Ctu+rnNE5Ctmr+ymBVgarElxnBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e873ad47d375d0572c86cf81f4f47d77b33f518802ade3c85be0d37b2861d632","last_reissued_at":"2026-07-17T01:22:17.454759Z","signature_status":"signed_v1","first_computed_at":"2026-07-17T01:22:17.454759Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CR","authors_text":"Blaine Nelson, Paul Kassianik, Yaron Singer","submitted_at":"2026-07-16T17:54:47Z","abstract_excerpt":"Security-agent evaluations commonly measure peak offensive capability under generous inference budgets, emphasizing vulnerability discovery, exploit development, penetration testing, and CTF completion. Such measurements are useful but incomplete: in operational security, every reasoning step, tool call, telemetry query, and enrichment request consumes budget. We evaluate language-model security agents through this cost-success lens on offensive Cybench challenges and defensive Splunk BOTS v1 investigation challenges. Instead of reporting only best-case success, we compare models at fixed cost"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.15263","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/2607.15263/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":"2607.15263","created_at":"2026-07-17T01:22:17.455193+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.15263v1","created_at":"2026-07-17T01:22:17.455193+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.15263","created_at":"2026-07-17T01:22:17.455193+00:00"},{"alias_kind":"pith_short_12","alias_value":"5BZ22R6TOXIF","created_at":"2026-07-17T01:22:17.455193+00:00"},{"alias_kind":"pith_short_16","alias_value":"5BZ22R6TOXIFOLEG","created_at":"2026-07-17T01:22:17.455193+00:00"},{"alias_kind":"pith_short_8","alias_value":"5BZ22R6T","created_at":"2026-07-17T01:22:17.455193+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6","json":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6.json","graph_json":"https://pith.science/api/pith-number/5BZ22R6TOXIFOLEGZ6A7J5D5O6/graph.json","events_json":"https://pith.science/api/pith-number/5BZ22R6TOXIFOLEGZ6A7J5D5O6/events.json","paper":"https://pith.science/paper/5BZ22R6T"},"agent_actions":{"view_html":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6","download_json":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6.json","view_paper":"https://pith.science/paper/5BZ22R6T","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.15263&json=true","fetch_graph":"https://pith.science/api/pith-number/5BZ22R6TOXIFOLEGZ6A7J5D5O6/graph.json","fetch_events":"https://pith.science/api/pith-number/5BZ22R6TOXIFOLEGZ6A7J5D5O6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6/action/storage_attestation","attest_author":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6/action/author_attestation","sign_citation":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6/action/citation_signature","submit_replication":"https://pith.science/pith/5BZ22R6TOXIFOLEGZ6A7J5D5O6/action/replication_record"}},"created_at":"2026-07-17T01:22:17.455193+00:00","updated_at":"2026-07-17T01:22:17.455193+00:00"}