{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:XWWVQBXM6G3Q7VFSDCW6KYT4WJ","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":"5ad5b5d483daf37292f905793bca551666fb7ee48440ce2189e8c074e30a0c7d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-12T13:30:06Z","title_canon_sha256":"af34f38dea61b3c98defedbfc9989ae02eb10ccc279706ba78a3f31047e2dfe8"},"schema_version":"1.0","source":{"id":"2411.07781","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.07781","created_at":"2026-07-05T09:34:28Z"},{"alias_kind":"arxiv_version","alias_value":"2411.07781v1","created_at":"2026-07-05T09:34:28Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.07781","created_at":"2026-07-05T09:34:28Z"},{"alias_kind":"pith_short_12","alias_value":"XWWVQBXM6G3Q","created_at":"2026-07-05T09:34:28Z"},{"alias_kind":"pith_short_16","alias_value":"XWWVQBXM6G3Q7VFS","created_at":"2026-07-05T09:34:28Z"},{"alias_kind":"pith_short_8","alias_value":"XWWVQBXM","created_at":"2026-07-05T09:34:28Z"}],"graph_snapshots":[{"event_id":"sha256:ccbf00596cb0571489e646d601d9e70621023847e53411345ea6c51f5f2cde79","target":"graph","created_at":"2026-07-05T09:34:28Z","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/2411.07781/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the rapidly increasing capabilities and adoption of code agents for AI-assisted coding, safety concerns, such as generating or executing risky code, have become significant barriers to the real-world deployment of these agents. To provide comprehensive and practical evaluations on the safety of code agents, we propose RedCode, a benchmark for risky code execution and generation: (1) RedCode-Exec provides challenging prompts that could lead to risky code execution, aiming to evaluate code agents' ability to recognize and handle unsafe code. We provide a total of 4,050 risky test cases in P","authors_text":"Andy Zhou, Bo Li, Chengquan Guo, Chulin Xie, Dawn Song, Xun Liu, Yi Zeng, Zinan Lin","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-12T13:30:06Z","title":"RedCode: Risky Code Execution and Generation Benchmark for Code Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.07781","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":"sha256:88b9044c15847ad7d941fb54b0e1f2106357bfb9541476c7e797a73350350837","target":"record","created_at":"2026-07-05T09:34:28Z","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":"5ad5b5d483daf37292f905793bca551666fb7ee48440ce2189e8c074e30a0c7d","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2024-11-12T13:30:06Z","title_canon_sha256":"af34f38dea61b3c98defedbfc9989ae02eb10ccc279706ba78a3f31047e2dfe8"},"schema_version":"1.0","source":{"id":"2411.07781","kind":"arxiv","version":1}},"canonical_sha256":"bdad5806ecf1b70fd4b218ade5627cb27e791cd8e1661e9269ed97a85e62f300","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bdad5806ecf1b70fd4b218ade5627cb27e791cd8e1661e9269ed97a85e62f300","first_computed_at":"2026-07-05T09:34:28.559889Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:34:28.559889Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BUi8nFHY36TSH8QbMyx/s3/a6KSuJl9X9OvEMXZ9Cw92KUvgDaM4PRnhxOP6reHfkpSmwDmGjJrcaySaUFJMAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:34:28.560343Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.07781","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:88b9044c15847ad7d941fb54b0e1f2106357bfb9541476c7e797a73350350837","sha256:ccbf00596cb0571489e646d601d9e70621023847e53411345ea6c51f5f2cde79"],"state_sha256":"359e7d83c9959773eb486e12e16a5ecb045c3ee2d046c8bdb73e6892a86a695d"}