{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:G6KZJHZHTVO2SOHKRQRV5TAB72","short_pith_number":"pith:G6KZJHZH","schema_version":"1.0","canonical_sha256":"3795949f279d5da938ea8c235ecc01fe99591306bd53c0f549b0f4d0efa53204","source":{"kind":"arxiv","id":"2607.03968","version":1},"attestation_state":"computed","paper":{"title":"Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Abhishek Kumar, Carsten Maple","submitted_at":"2026-07-04T17:57:05Z","abstract_excerpt":"Large language models are increasingly deployed as IDE-integrated coding agents that decompose tasks, generate and edit files, run code, and refine outputs over many turns. Yet their safety is still often evaluated as if they were chatbots: one harmful prompt, one response, judged in isolation. We introduce workflow-level jailbreak construction, a failure mode in which a harmful objective is assembled across ordinary stages of a software-development workflow rather than generated through a single direct prompt. Using GitHub Copilot in Visual Studio Code, we study four closed-weight backends: C"},"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":"2607.03968","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2026-07-04T17:57:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c75653746552765a9a8563dd4bcc51551a3122b2def5e368737cd11adbdcf3f5","abstract_canon_sha256":"0b544f8575cfdf3ddbe275989e13c6afc86e09420f45796ca10057f59ba59479"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:18:50.240065Z","signature_b64":"a6KLQkyCm2v+78aFTfV1zco0Tk9SAzb2o2GpmVrI4fK+VhxG/hngRI+XHX17njlIPnxHEcmcDdsFvg6lFoneBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3795949f279d5da938ea8c235ecc01fe99591306bd53c0f549b0f4d0efa53204","last_reissued_at":"2026-07-07T02:18:50.239418Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:18:50.239418Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Refused in Chat, Written in Code: Workflow-Level Jailbreak Construction in IDE Coding Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.SE","authors_text":"Abhishek Kumar, Carsten Maple","submitted_at":"2026-07-04T17:57:05Z","abstract_excerpt":"Large language models are increasingly deployed as IDE-integrated coding agents that decompose tasks, generate and edit files, run code, and refine outputs over many turns. Yet their safety is still often evaluated as if they were chatbots: one harmful prompt, one response, judged in isolation. We introduce workflow-level jailbreak construction, a failure mode in which a harmful objective is assembled across ordinary stages of a software-development workflow rather than generated through a single direct prompt. Using GitHub Copilot in Visual Studio Code, we study four closed-weight backends: C"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.03968","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/2607.03968/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":"2607.03968","created_at":"2026-07-07T02:18:50.239505+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.03968v1","created_at":"2026-07-07T02:18:50.239505+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.03968","created_at":"2026-07-07T02:18:50.239505+00:00"},{"alias_kind":"pith_short_12","alias_value":"G6KZJHZHTVO2","created_at":"2026-07-07T02:18:50.239505+00:00"},{"alias_kind":"pith_short_16","alias_value":"G6KZJHZHTVO2SOHK","created_at":"2026-07-07T02:18:50.239505+00:00"},{"alias_kind":"pith_short_8","alias_value":"G6KZJHZH","created_at":"2026-07-07T02:18:50.239505+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/G6KZJHZHTVO2SOHKRQRV5TAB72","json":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72.json","graph_json":"https://pith.science/api/pith-number/G6KZJHZHTVO2SOHKRQRV5TAB72/graph.json","events_json":"https://pith.science/api/pith-number/G6KZJHZHTVO2SOHKRQRV5TAB72/events.json","paper":"https://pith.science/paper/G6KZJHZH"},"agent_actions":{"view_html":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72","download_json":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72.json","view_paper":"https://pith.science/paper/G6KZJHZH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.03968&json=true","fetch_graph":"https://pith.science/api/pith-number/G6KZJHZHTVO2SOHKRQRV5TAB72/graph.json","fetch_events":"https://pith.science/api/pith-number/G6KZJHZHTVO2SOHKRQRV5TAB72/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72/action/storage_attestation","attest_author":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72/action/author_attestation","sign_citation":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72/action/citation_signature","submit_replication":"https://pith.science/pith/G6KZJHZHTVO2SOHKRQRV5TAB72/action/replication_record"}},"created_at":"2026-07-07T02:18:50.239505+00:00","updated_at":"2026-07-07T02:18:50.239505+00:00"}