{"paper":{"title":"\"Tab, Tab, Bug\": Security Pitfalls of Next Edit Suggestions in AI-Integrated IDEs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Next Edit Suggestions in AI IDEs expand context retrieval in ways that enable poisoning attacks.","cross_cats":["cs.HC"],"primary_cat":"cs.CR","authors_text":"Hao Chen, Peng Chen, Tian Dong, Xinyu Wang, Yixuan Tang, Yunlong Lyu, Zhiqiang Dong","submitted_at":"2026-02-06T15:06:36Z","abstract_excerpt":"Modern AI-integrated IDEs are shifting from passive code completion to proactive Next Edit Suggestions (NES). Unlike traditional autocompletion, NES is designed to construct a richer context from both recent user interactions and the broader codebase to suggest multi-line, cross-line, or even cross-file modifications. This evolution significantly streamlines the programming workflow into a tab-by-tab interaction and enhances developer productivity. Consequently, NES introduces a more complex context retrieval mechanism and sophisticated interaction patterns. However, existing studies focus alm"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"NES is susceptible to context poisoning and is sensitive to transactional edits and human-IDE interactions; developers show a general lack of awareness of these security pitfalls.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The in-lab attack scenarios and survey responses accurately reflect real-world attacker capabilities and developer behavior without significant selection or reporting bias.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"NES systems in AI IDEs expand attack surfaces via context poisoning from imperceptible actions and global codebase retrieval, with professional developers largely unaware of the risks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Next Edit Suggestions in AI IDEs expand context retrieval in ways that enable poisoning attacks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"ac1e22452980d68bf7fbab94f7364b63e7b4eb2456d2e3370cf70aff66f63ee0"},"source":{"id":"2602.06759","kind":"arxiv","version":2},"verdict":{"id":"84a6dd57-2917-4fb9-85a5-b2d0993a9e15","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-16T06:50:18.687412Z","strongest_claim":"NES is susceptible to context poisoning and is sensitive to transactional edits and human-IDE interactions; developers show a general lack of awareness of these security pitfalls.","one_line_summary":"NES systems in AI IDEs expand attack surfaces via context poisoning from imperceptible actions and global codebase retrieval, with professional developers largely unaware of the risks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The in-lab attack scenarios and survey responses accurately reflect real-world attacker capabilities and developer behavior without significant selection or reporting bias.","pith_extraction_headline":"Next Edit Suggestions in AI IDEs expand context retrieval in ways that enable poisoning attacks."},"references":{"count":49,"sample":[{"doi":"10.1145/2702123.2702322","year":2015,"title":"Brock Kirwan, Jeffrey L","work_id":"34d25f97-0979-4a3c-a67f-be04f2ea127a","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Efficient training of language models to fill in the middle","work_id":"54afe4f8-4d93-4829-99ae-2a27143a9641","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Purple llama CyberSecEval : A secure coding benchmark for language models","work_id":"45b8079b-5204-450f-8024-f3a8142583a9","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2021,"title":"Evaluating Large Language Models Trained on Code","work_id":"042493e9-b26f-4b4e-bbde-382072ca9b08","ref_index":4,"cited_arxiv_id":"2107.03374","is_internal_anchor":true},{"doi":"","year":2025,"title":"An efficient and adaptive next edit suggestion framework with zero human instructions in ides, 2025","work_id":"a015fdac-6db4-45bd-83f9-18e050a5e437","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":49,"snapshot_sha256":"3b0d2e38dc653df8bc023863dd1b1850bd337adb6a8c6115cf4bf90070ecd3a1","internal_anchors":7},"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"}