{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:XQXLJV74NXCYXWH36VKAM4WBRE","short_pith_number":"pith:XQXLJV74","canonical_record":{"source":{"id":"2606.07867","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T21:58:19Z","cross_cats_sorted":[],"title_canon_sha256":"97d490be9f755cbead2d167f98ead0da4c6ab8d6837754f145f1e53a4c967a53","abstract_canon_sha256":"1077ed1f026ef71bd9aca4d5702111e37af6bd48bf23324324f3a078f99cef3c"},"schema_version":"1.0"},"canonical_sha256":"bc2eb4d7fc6dc58bd8fbf5540672c18904d9e8703d48f85319c5d7fc759b2df2","source":{"kind":"arxiv","id":"2606.07867","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.07867","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"arxiv_version","alias_value":"2606.07867v1","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.07867","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"pith_short_12","alias_value":"XQXLJV74NXCY","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"pith_short_16","alias_value":"XQXLJV74NXCYXWH3","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"pith_short_8","alias_value":"XQXLJV74","created_at":"2026-06-09T01:04:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:XQXLJV74NXCYXWH36VKAM4WBRE","target":"record","payload":{"canonical_record":{"source":{"id":"2606.07867","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T21:58:19Z","cross_cats_sorted":[],"title_canon_sha256":"97d490be9f755cbead2d167f98ead0da4c6ab8d6837754f145f1e53a4c967a53","abstract_canon_sha256":"1077ed1f026ef71bd9aca4d5702111e37af6bd48bf23324324f3a078f99cef3c"},"schema_version":"1.0"},"canonical_sha256":"bc2eb4d7fc6dc58bd8fbf5540672c18904d9e8703d48f85319c5d7fc759b2df2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-09T01:04:54.149634Z","signature_b64":"L1lfwP3wVOTjPfi4mlbvEjmCbD1XpXQA+zVPJyJeYMXyg3sSDzBXWwIfDca5I9W7G+gHewIgl/2wxSw7MihQBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bc2eb4d7fc6dc58bd8fbf5540672c18904d9e8703d48f85319c5d7fc759b2df2","last_reissued_at":"2026-06-09T01:04:54.149222Z","signature_status":"signed_v1","first_computed_at":"2026-06-09T01:04:54.149222Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.07867","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":"2026-06-09T01:04:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Dmys0hEkELajGdsrAcjDZTqdBMUITaSVRsj6W15Zptwua9jOojnuSXTlOm/swxLL9rTm2b865FSG3EExeXhKBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T19:54:02.239149Z"},"content_sha256":"54e8b0065dcba5dcab3aa24e99609826f163257a7bcca4c35473f04c5a9202a6","schema_version":"1.0","event_id":"sha256:54e8b0065dcba5dcab3aa24e99609826f163257a7bcca4c35473f04c5a9202a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:XQXLJV74NXCYXWH36VKAM4WBRE","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Cold-Start Safety Gap in LLM Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chung-En Sun, Linbo Liu, Tsui-Wei Weng","submitted_at":"2026-06-05T21:58:19Z","abstract_excerpt":"Are tool-calling LLM agents equally safe throughout a conversation? We discover they are not: agents are most vulnerable at the very start of a session and become substantially safer after a few regular agentic tasks -- a phenomenon we term the cold-start safety gap. To study this systematically, we introduce Safety Over Depth for Agents (SODA), a benchmark that controls how many regular agentic tasks the agent completes before encountering a safety threat, supporting up to 20 preceding tasks. Evaluating 7 models from 4 families, safety improves by 9--52% as the number of preceding regular age"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.07867","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/2606.07867/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":"2026-06-09T01:04:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iTKIJcJceLHryx8t2xGisnmppP37nTT+k4HohWR2YACOXolfC/IYaCN/QPBd7U1uaG7Eqw/slsyvnU+aJaEQCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-11T19:54:02.239960Z"},"content_sha256":"9c5f5b96b6ae0287de9a873c32f33402a6223f44e59c9f37c8c8fb2a8b1bd979","schema_version":"1.0","event_id":"sha256:9c5f5b96b6ae0287de9a873c32f33402a6223f44e59c9f37c8c8fb2a8b1bd979"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XQXLJV74NXCYXWH36VKAM4WBRE/bundle.json","state_url":"https://pith.science/pith/XQXLJV74NXCYXWH36VKAM4WBRE/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XQXLJV74NXCYXWH36VKAM4WBRE/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":"2026-06-11T19:54:02Z","links":{"resolver":"https://pith.science/pith/XQXLJV74NXCYXWH36VKAM4WBRE","bundle":"https://pith.science/pith/XQXLJV74NXCYXWH36VKAM4WBRE/bundle.json","state":"https://pith.science/pith/XQXLJV74NXCYXWH36VKAM4WBRE/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XQXLJV74NXCYXWH36VKAM4WBRE/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XQXLJV74NXCYXWH36VKAM4WBRE","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":"1077ed1f026ef71bd9aca4d5702111e37af6bd48bf23324324f3a078f99cef3c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T21:58:19Z","title_canon_sha256":"97d490be9f755cbead2d167f98ead0da4c6ab8d6837754f145f1e53a4c967a53"},"schema_version":"1.0","source":{"id":"2606.07867","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.07867","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"arxiv_version","alias_value":"2606.07867v1","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.07867","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"pith_short_12","alias_value":"XQXLJV74NXCY","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"pith_short_16","alias_value":"XQXLJV74NXCYXWH3","created_at":"2026-06-09T01:04:54Z"},{"alias_kind":"pith_short_8","alias_value":"XQXLJV74","created_at":"2026-06-09T01:04:54Z"}],"graph_snapshots":[{"event_id":"sha256:9c5f5b96b6ae0287de9a873c32f33402a6223f44e59c9f37c8c8fb2a8b1bd979","target":"graph","created_at":"2026-06-09T01:04:54Z","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/2606.07867/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Are tool-calling LLM agents equally safe throughout a conversation? We discover they are not: agents are most vulnerable at the very start of a session and become substantially safer after a few regular agentic tasks -- a phenomenon we term the cold-start safety gap. To study this systematically, we introduce Safety Over Depth for Agents (SODA), a benchmark that controls how many regular agentic tasks the agent completes before encountering a safety threat, supporting up to 20 preceding tasks. Evaluating 7 models from 4 families, safety improves by 9--52% as the number of preceding regular age","authors_text":"Chung-En Sun, Linbo Liu, Tsui-Wei Weng","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T21:58:19Z","title":"The Cold-Start Safety Gap in LLM Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.07867","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":"sha256:54e8b0065dcba5dcab3aa24e99609826f163257a7bcca4c35473f04c5a9202a6","target":"record","created_at":"2026-06-09T01:04:54Z","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":"1077ed1f026ef71bd9aca4d5702111e37af6bd48bf23324324f3a078f99cef3c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-05T21:58:19Z","title_canon_sha256":"97d490be9f755cbead2d167f98ead0da4c6ab8d6837754f145f1e53a4c967a53"},"schema_version":"1.0","source":{"id":"2606.07867","kind":"arxiv","version":1}},"canonical_sha256":"bc2eb4d7fc6dc58bd8fbf5540672c18904d9e8703d48f85319c5d7fc759b2df2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bc2eb4d7fc6dc58bd8fbf5540672c18904d9e8703d48f85319c5d7fc759b2df2","first_computed_at":"2026-06-09T01:04:54.149222Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-09T01:04:54.149222Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L1lfwP3wVOTjPfi4mlbvEjmCbD1XpXQA+zVPJyJeYMXyg3sSDzBXWwIfDca5I9W7G+gHewIgl/2wxSw7MihQBw==","signature_status":"signed_v1","signed_at":"2026-06-09T01:04:54.149634Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.07867","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54e8b0065dcba5dcab3aa24e99609826f163257a7bcca4c35473f04c5a9202a6","sha256:9c5f5b96b6ae0287de9a873c32f33402a6223f44e59c9f37c8c8fb2a8b1bd979"],"state_sha256":"db99365279363060244bdf16bc5bb6c00b212bed427fb0dca20edb22b77ddd60"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CgD/K5OxcCizMYh44b5a8ZK1NFqLB0y7e+dopmcFRZUnxQY/XummqhsKAMnJV8OF7s8afBzjatrSoRVGDY1fCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-11T19:54:02.247428Z","bundle_sha256":"4d2d3744217009c7315098fef04049f59562e80fb2e6225be8612121301efe07"}}