[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-github-answers-400-more-commits-with-more-agents":10,"sections":41},{"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":30,"tags":31,"sources":36,"feedback":40,"feedback_at":22,"cost_usd":40,"total_tokens":40},12418,"github-answers-400-more-commits-with-more-agents","GitHub Answers 400% More Commits With More Agents","GitHub says review, not code writing, is the new bottleneck, and its guide pitches Copilot agents for planning, building and verifying.","GitHub says developers are writing far more code than they can comfortably review, and it wants Copilot agents to handle the slack at every stage of the job.\n\nThe company said it logged more than 1.2 billion commits in one week this October, over 400% more than a year earlier and more than all of 2025. It paired that with a Faros AI study, which found pull requests spent 441% longer in review when teams went from low to high AI adoption. This piece covers three stages of the workflow GitHub describes: Plan, Build and Verify.\n\nIn Plan, Copilot works inside Slack or Teams, turns agreed requirements into GitHub Issues with acceptance criteria, and uses GitHub Agentic Workflows to label issues and estimate complexity. Low-complexity tasks can go straight to Copilot. In Build, sessions carry across devices, run in cloud sandboxes and get isolated copies of the repository. Shared memory and a feature called Copilot Dreaming keep project conventions current, and Copilot Fusion lets you pick models and reasoning depth.\n\nVerify is where the guide is most candid. It describes a pull request labeled as a small refactor that touches 47 files, and calls that review queue the 441% in practice. The fixes include stacked pull requests, GitHub Code Quality, Copilot code review with adjustable depth, AI Scan for languages beyond CodeQL's coverage, and AI secret detection. A package firewall, in public preview for npm, can block downloads that break policy. Agentic Autofix proposes repairs for security issues, while Agent Merge addresses failing checks, review comments and merge conflicts.\n\nCommit counts measure activity, not output, and GitHub's own framing admits as much by asking whether anyone feels 400% more productive. The sharpest advice here is also the least automated: smaller, stacked pull requests make a human reviewer's job easier. Most of the rest adds machine reviewers in front of the human one. Whether that shortens the queue or just hands people larger piles of agent-polished diffs is a question the guide leaves unanswered.","[\"github\",\"copilot\",\"code-review\",\"ai-agents\"]","2026-11-06T22:15:59.000Z","2026-11-08T01:13:17.097Z","2026-11-08T01:13:17.388Z","published",null,[24],{"id":25,"reviewer":26,"round":27,"reason":28,"status":29},"editor-r1","editor",1,"Delete the meta-text line \"The excerpt covers the first three.\" and either drop the Observe stage or say plainly that this piece covers only Plan, Build and Verify without referring to an excerpt. Also correct the Verify description: Agentic Autofix proposes repairs for security issues, while Agent Merge is the one that addresses failing checks, review comments and merge conflicts.","resolved","dev-tools",[32,33,34,35],"github","copilot","code-review","ai-agents",[37],{"name":38,"url":39},"GitHub Blog","https:\u002F\u002Fgithub.blog\u002Fai-and-ml\u002Fgithub-copilot\u002Fbeyond-code-generation-a-developers-guide-to-agentic-engineering\u002F",0,{"sections":42},[43,48,53,58,63,68,73,78,83,88,91,96,101,106],{"name":44,"slug":45,"count":46,"latest_published_at":47},"AI","ai",7110,"2026-11-08T14:34:09.000Z",{"name":49,"slug":50,"count":51,"latest_published_at":52},"Security","security",1099,"2026-11-08T15:24:01.000Z",{"name":54,"slug":55,"count":56,"latest_published_at":57},"Deals","deals",620,"2026-11-08T08:00:00.000Z",{"name":59,"slug":60,"count":61,"latest_published_at":62},"Policy","policy",604,"2026-11-07T15:48:26.000Z",{"name":64,"slug":65,"count":66,"latest_published_at":67},"Consumer Tech","consumer-tech",336,"2026-11-08T21:00:00.000Z",{"name":69,"slug":70,"count":71,"latest_published_at":72},"Hardware","hardware",316,"2026-11-08T14:36:39.000Z",{"name":74,"slug":75,"count":76,"latest_published_at":77},"Science","science",222,"2026-11-07T16:19:17.000Z",{"name":79,"slug":80,"count":81,"latest_published_at":82},"Software","software",154,"2026-11-08T15:42:52.000Z",{"name":84,"slug":85,"count":86,"latest_published_at":87},"Startups","startups",150,"2026-11-06T16:50:01.000Z",{"name":89,"slug":30,"count":90,"latest_published_at":18},"Dev Tools",125,{"name":92,"slug":93,"count":94,"latest_published_at":95},"General","general",111,"2026-11-08T21:27:29.000Z",{"name":97,"slug":98,"count":99,"latest_published_at":100},"Gaming","gaming",94,"2026-11-08T20:52:05.000Z",{"name":102,"slug":103,"count":104,"latest_published_at":105},"Reviews","reviews",38,"2026-10-20T12:00:00.000Z",{"name":107,"slug":108,"count":109,"latest_published_at":110},"How-To","how-to",8,"2026-10-05T09:00:00.000Z"]