{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:2KDVJ77EEHFTTKYQVALBO6GAXD","short_pith_number":"pith:2KDVJ77E","canonical_record":{"source":{"id":"2504.00018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-03-27T19:56:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a67b7465551075c677090b8163ef7450119f9ce5b49bccf4c7d9fd90cd8e434b","abstract_canon_sha256":"e37ebdaa5b815ebf9ade2029352e1fe487c7030ce77b3ad8974b2c4bdb5b4b07"},"schema_version":"1.0"},"canonical_sha256":"d28754ffe421cb39ab10a8161778c0b8c863bca5dcec0fc6c487bf7e68172349","source":{"kind":"arxiv","id":"2504.00018","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.00018","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"arxiv_version","alias_value":"2504.00018v1","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.00018","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"pith_short_12","alias_value":"2KDVJ77EEHFT","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"pith_short_16","alias_value":"2KDVJ77EEHFTTKYQ","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"pith_short_8","alias_value":"2KDVJ77E","created_at":"2026-07-05T10:42:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:2KDVJ77EEHFTTKYQVALBO6GAXD","target":"record","payload":{"canonical_record":{"source":{"id":"2504.00018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-03-27T19:56:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"a67b7465551075c677090b8163ef7450119f9ce5b49bccf4c7d9fd90cd8e434b","abstract_canon_sha256":"e37ebdaa5b815ebf9ade2029352e1fe487c7030ce77b3ad8974b2c4bdb5b4b07"},"schema_version":"1.0"},"canonical_sha256":"d28754ffe421cb39ab10a8161778c0b8c863bca5dcec0fc6c487bf7e68172349","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:42:18.453466Z","signature_b64":"+UteIAIeA2sQPDv7U+bOrzVHU3wAsheXT/dx7MPGyFlzyoZa6JF8FgDZaZqRsLj1OmiwYgX3RsClaf2z+5qBDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d28754ffe421cb39ab10a8161778c0b8c863bca5dcec0fc6c487bf7e68172349","last_reissued_at":"2026-07-05T10:42:18.452963Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:42:18.452963Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.00018","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:42:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nIRipNwKdIGmMqxrHVrjsNvVcRE1vf4lQKhB2TqgeBhRARvNrAf0yKAJ6SmskuIx5ql+Azs4TTI8m/X92pORBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T18:40:57.754629Z"},"content_sha256":"501ba92e6d0b711531e134f9638542d8caf6683b0fb82592cbdc162aae18f184","schema_version":"1.0","event_id":"sha256:501ba92e6d0b711531e134f9638542d8caf6683b0fb82592cbdc162aae18f184"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:2KDVJ77EEHFTTKYQVALBO6GAXD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SandboxEval: Towards Securing Test Environment for Untrusted Code","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Brett Weir, Jesse Hostetler, Nick Judd, Rafiqul Rabin, Sean McGregor","submitted_at":"2025-03-27T19:56:00Z","abstract_excerpt":"While large language models (LLMs) are powerful assistants in programming tasks, they may also produce malicious code. Testing LLM-generated code therefore poses significant risks to assessment infrastructure tasked with executing untrusted code. To address these risks, this work focuses on evaluating the security and confidentiality properties of test environments, reducing the risk that LLM-generated code may compromise the assessment infrastructure. We introduce SandboxEval, a test suite featuring manually crafted test cases that simulate real-world safety scenarios for LLM assessment envir"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.00018","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/2504.00018/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:42:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sZxJzTYoVYAOZh3GrSUARGy5z3BolFDZPYALhZh00KyvZA5dkA7lDVKw4BZwJI0naHsxj3purklkUlQdb+5BCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-24T18:40:57.755005Z"},"content_sha256":"8b8ea23bf707534b7305378d9f495ecf059985a7aadf9b001da75b8a6cda5e10","schema_version":"1.0","event_id":"sha256:8b8ea23bf707534b7305378d9f495ecf059985a7aadf9b001da75b8a6cda5e10"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2KDVJ77EEHFTTKYQVALBO6GAXD/bundle.json","state_url":"https://pith.science/pith/2KDVJ77EEHFTTKYQVALBO6GAXD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2KDVJ77EEHFTTKYQVALBO6GAXD/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-24T18:40:57Z","links":{"resolver":"https://pith.science/pith/2KDVJ77EEHFTTKYQVALBO6GAXD","bundle":"https://pith.science/pith/2KDVJ77EEHFTTKYQVALBO6GAXD/bundle.json","state":"https://pith.science/pith/2KDVJ77EEHFTTKYQVALBO6GAXD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2KDVJ77EEHFTTKYQVALBO6GAXD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:2KDVJ77EEHFTTKYQVALBO6GAXD","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":"e37ebdaa5b815ebf9ade2029352e1fe487c7030ce77b3ad8974b2c4bdb5b4b07","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-03-27T19:56:00Z","title_canon_sha256":"a67b7465551075c677090b8163ef7450119f9ce5b49bccf4c7d9fd90cd8e434b"},"schema_version":"1.0","source":{"id":"2504.00018","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.00018","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"arxiv_version","alias_value":"2504.00018v1","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.00018","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"pith_short_12","alias_value":"2KDVJ77EEHFT","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"pith_short_16","alias_value":"2KDVJ77EEHFTTKYQ","created_at":"2026-07-05T10:42:18Z"},{"alias_kind":"pith_short_8","alias_value":"2KDVJ77E","created_at":"2026-07-05T10:42:18Z"}],"graph_snapshots":[{"event_id":"sha256:8b8ea23bf707534b7305378d9f495ecf059985a7aadf9b001da75b8a6cda5e10","target":"graph","created_at":"2026-07-05T10:42:18Z","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/2504.00018/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While large language models (LLMs) are powerful assistants in programming tasks, they may also produce malicious code. Testing LLM-generated code therefore poses significant risks to assessment infrastructure tasked with executing untrusted code. To address these risks, this work focuses on evaluating the security and confidentiality properties of test environments, reducing the risk that LLM-generated code may compromise the assessment infrastructure. We introduce SandboxEval, a test suite featuring manually crafted test cases that simulate real-world safety scenarios for LLM assessment envir","authors_text":"Brett Weir, Jesse Hostetler, Nick Judd, Rafiqul Rabin, Sean McGregor","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-03-27T19:56:00Z","title":"SandboxEval: Towards Securing Test Environment for Untrusted Code"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.00018","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":"sha256:501ba92e6d0b711531e134f9638542d8caf6683b0fb82592cbdc162aae18f184","target":"record","created_at":"2026-07-05T10:42:18Z","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":"e37ebdaa5b815ebf9ade2029352e1fe487c7030ce77b3ad8974b2c4bdb5b4b07","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2025-03-27T19:56:00Z","title_canon_sha256":"a67b7465551075c677090b8163ef7450119f9ce5b49bccf4c7d9fd90cd8e434b"},"schema_version":"1.0","source":{"id":"2504.00018","kind":"arxiv","version":1}},"canonical_sha256":"d28754ffe421cb39ab10a8161778c0b8c863bca5dcec0fc6c487bf7e68172349","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d28754ffe421cb39ab10a8161778c0b8c863bca5dcec0fc6c487bf7e68172349","first_computed_at":"2026-07-05T10:42:18.452963Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:42:18.452963Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+UteIAIeA2sQPDv7U+bOrzVHU3wAsheXT/dx7MPGyFlzyoZa6JF8FgDZaZqRsLj1OmiwYgX3RsClaf2z+5qBDA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:42:18.453466Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.00018","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:501ba92e6d0b711531e134f9638542d8caf6683b0fb82592cbdc162aae18f184","sha256:8b8ea23bf707534b7305378d9f495ecf059985a7aadf9b001da75b8a6cda5e10"],"state_sha256":"67ac82a52f66a100a6a9c5e250c2b822ba1b820f3bf87faceb92e3f413428725"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B41kmjj/SF7nemnUujXXvIVhA37/fBsO3us9QR2Q0YG3l109ozHkBdy9dVL+BgH2Qre1dKhrAimc2KwkJVpmAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-24T18:40:57.757223Z","bundle_sha256":"9cd4b7e2fc261c95265963d6104faaf51e20042ac875ce85be2ae91d7538d610"}}