{"schema_version":"papers/paper-detail-v1","title":"ExploitBench: A Capability Ladder Benchmark for LLM Cybersecurity Agents","surface":"papers","opportunity_kernel":{"paper_id":"dd991f5b-9316-42d4-8e37-a585aa6ab4dc","title":"ExploitBench: A Capability Ladder Benchmark for LLM Cybersecurity Agents","authors":["Seunghyun Lee","David Brumley"],"arxiv_id":"2605.14153v1","doi":null,"published_at":"2026-05-13T22:08:05.000Z","score_object":{"overall":{"value":7,"scale":"0-10","confidence":0.85,"confidence_reason":"Backfilled from persisted papers.viability_score.","model_version":"phase0-web-fallback-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-15T20:12:56.418Z","fresh_until":"2026-06-14T20:12:56.418Z","is_stale":false,"source_count":1,"missingness":[]},"technical":{"value":0,"scale":"0-10","confidence":0.15,"confidence_reason":"No canonical technical score row was available yet; marked low confidence.","model_version":"phase0-web-fallback-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-15T20:12:56.418Z","fresh_until":"2026-05-29T20:12:56.418Z","is_stale":true,"source_count":0,"missingness":["paper_score_objects.technical"]},"commercial":{"value":4,"scale":"0-10","confidence":0.7,"confidence_reason":"Backfilled from persisted commercial_flags and repo availability.","model_version":"phase0-web-fallback-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-15T20:12:56.418Z","fresh_until":"2026-06-14T20:12:56.418Z","is_stale":false,"source_count":1,"missingness":[]},"market":{"value":0,"scale":"0-10","confidence":0.15,"confidence_reason":"No canonical market score row was available yet; marked low confidence.","model_version":"phase0-web-fallback-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-15T20:12:56.418Z","fresh_until":"2026-05-29T20:12:56.418Z","is_stale":true,"source_count":0,"missingness":["paper_score_objects.market"]},"team":{"value":0,"scale":"0-10","confidence":0.15,"confidence_reason":"No canonical team score row was available yet; marked low confidence.","model_version":"phase0-web-fallback-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-15T20:12:56.418Z","fresh_until":"2026-05-29T20:12:56.418Z","is_stale":true,"source_count":0,"missingness":["paper_score_objects.team"]},"methodology":{"value":0,"scale":"0-10","confidence":0.15,"confidence_reason":"No canonical methodology score row was available yet; marked low confidence.","model_version":"phase0-web-fallback-v1","pipeline_version":"phase0-kernel-v1","computed_at":"2026-05-15T20:12:56.418Z","fresh_until":"2026-06-14T20:12:56.418Z","is_stale":false,"source_count":0,"missingness":["paper_score_objects.methodology"]}},"evidence_receipt":{"freshness":"fresh","proof_status":"unverified","repo_status":"missing","references_count":0,"source_count":0,"coverage":0,"missingness":["paper_evidence_receipts.references_count","paper_evidence_receipts.coverage"],"unresolved_unknowns":["Canonical evidence receipt has not been materialized yet."],"last_verification_at":"2026-05-15T20:12:56.418Z"},"lineage_hash":"2b0fdcad95e696aed251638fd0eb05fb6fa372be0b991e8a44f4d7425d18e734"},"distribution":null,"replication_evidence":[],"author_dna":[]}