[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-the-coding-agent-bug-you-blamed-on-the-ai-was-probably-the-wrapper":10,"sections":34},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":29,"feedback":33,"feedback_at":22,"cost_usd":33,"total_tokens":33},3867,"the-coding-agent-bug-you-blamed-on-the-ai-was-probably-the-wrapper","The Coding Agent Bug You Blamed on the AI Was Probably the Wrapper","A new longitudinal study finds that scaffolding updates - not the underlying model - are the most likely culprit when coding agents suddenly get worse.","Coding agent quality regressions are being misdiagnosed, and a new study has the receipts.\n\nResearchers ran the first controlled longitudinal study isolating scaffolding as a variable in coding agent performance. Instead of swapping models and holding the scaffolding constant - the standard approach - they did the opposite: fixed the model and cycled through 35 sequential releases of the Qwen Code CLI, testing each against 50 standardized software engineering tasks from SWE-bench Verified. The five open-source scaffoldings examined - Codex, Qwen Code, Gemini, OpenCode, and OpenHands - were releasing at rates exceeding two versions per day, accumulating thousands of reported issues within months of launch.\n\nThat release velocity matters because scaffolding is doing more work than most developers realize. It sits between the developer and the model, handling system prompts, tool execution, context management, and iterative reasoning loops. Change any of those components and you change what the model actually does - even if the model itself is untouched. The study traced specific quality shifts directly to individual pull requests, which is about as granular as accountability gets.\n\nThe practical implication is uncomfortable for the current blame-the-model reflex: when a coding agent suddenly performs worse after an update, the first suspect should be the scaffolding layer, not the LLM. That distinction matters more as enterprises standardize on specific model versions while their scaffolding continues to churn underneath.\n\nThis is also a quiet indictment of how fast the tooling ecosystem is moving - shipping two releases a day sounds like momentum until a regression ships with them.","[\"ai\",\"dev-tools\",\"llm\",\"software-engineering\"]","2026-07-07T04:00:00.000Z","2026-07-07T10:25:21.898Z","2026-07-07T10:25:24.863Z","published",null,[],"ai",[24,26,27,28],"dev-tools","llm","software-engineering",[30],{"name":31,"url":32},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2607.03691",0,{"sections":35},[36,40,45,50,55,60,65,70,75,78,83,87,92,97],{"name":37,"slug":24,"count":38,"latest_published_at":39},"AI",2590,"2026-07-16T04:00:00.000Z",{"name":41,"slug":42,"count":43,"latest_published_at":44},"Security","security",294,"2026-07-15T19:59:48.000Z",{"name":46,"slug":47,"count":48,"latest_published_at":49},"Deals","deals",179,"2026-06-29T20:02:07.000Z",{"name":51,"slug":52,"count":53,"latest_published_at":54},"Policy","policy",158,"2026-07-16T00:02:48.000Z",{"name":56,"slug":57,"count":58,"latest_published_at":59},"Hardware","hardware",122,"2026-07-14T19:46:26.000Z",{"name":61,"slug":62,"count":63,"latest_published_at":64},"Consumer Tech","consumer-tech",93,"2026-07-13T13:20:48.000Z",{"name":66,"slug":67,"count":68,"latest_published_at":69},"Software","software",70,"2026-07-13T19:52:25.000Z",{"name":71,"slug":72,"count":73,"latest_published_at":74},"Science","science",66,"2026-07-10T10:29:37.000Z",{"name":76,"slug":26,"count":77,"latest_published_at":18},"Dev Tools",59,{"name":79,"slug":80,"count":81,"latest_published_at":82},"Gaming","gaming",41,"2026-07-09T04:00:00.000Z",{"name":84,"slug":85,"count":81,"latest_published_at":86},"Startups","startups","2026-06-29T20:55:50.000Z",{"name":88,"slug":89,"count":90,"latest_published_at":91},"General","general",29,"2026-07-10T22:28:58.000Z",{"name":93,"slug":94,"count":95,"latest_published_at":96},"Reviews","reviews",20,"2026-06-24T12:00:01.000Z",{"name":98,"slug":99,"count":100,"latest_published_at":101},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]