[{"data":1,"prerenderedAt":74},["ShallowReactive",2],{"term-a\u002Fauto-mode":3,"related-a\u002Fauto-mode":63},{"id":4,"title":5,"acronym":6,"body":7,"category":45,"description":46,"difficulty":47,"extension":48,"letter":49,"meta":50,"navigation":51,"path":52,"related":53,"seo":57,"sitemap":58,"stem":61,"subcategory":6,"__hash__":62},"terms\u002Fterms\u002Fa\u002Fauto-mode.md","Auto Mode",null,{"type":8,"value":9,"toc":38},"minimark",[10,15,19,23,31,35],[11,12,14],"h2",{"id":13},"eli5-the-vibe-check","ELI5 — The Vibe Check",[16,17,18],"p",{},"Auto mode is Claude Code's permission classifier — the middle ground between confirm-everything and YOLO mode. Instead of asking you before every file edit, it reads the situation and decides when to just do it versus when to tap you on the shoulder. It's like having an intern who's earned enough trust to handle the obvious stuff without a manager sign-off.",[11,20,22],{"id":21},"real-talk","Real Talk",[16,24,25,26,30],{},"In Claude Code, auto mode uses a built-in permission model that classifies actions by risk level. Low-risk operations (reading files, running safe scripts) proceed without confirmation. Higher-risk ones (deleting files, running network requests) still prompt. It sits between the fully interactive default mode and the nuclear ",[27,28,29],"code",{},"--dangerously-skip-permissions"," flag. Most developers land here after the first day of clicking \"yes\" to everything.",[11,32,34],{"id":33},"when-youll-hear-this","When You'll Hear This",[16,36,37],{},"\"Switch to auto mode — I trust it for this refactor.\" \u002F \"Auto mode tried to rm -rf something, so maybe we dial it back.\"",{"title":39,"searchDepth":40,"depth":40,"links":41},"",2,[42,43,44],{"id":13,"depth":40,"text":14},{"id":21,"depth":40,"text":22},{"id":33,"depth":40,"text":34},"vibecoding","Auto mode is Claude Code's permission classifier — the middle ground between confirm-everything and YOLO mode.","beginner","md","a",{},true,"\u002Fterms\u002Fa\u002Fauto-mode",[54,55,56],"Vibe Coding","Claude Code","Permission Mode",{"title":5,"description":46},{"changefreq":59,"priority":60},"weekly",0.7,"terms\u002Fa\u002Fauto-mode","f97Sv7D76gmRNifSJ4aLpiBSHqCUZzSAihMhYHJJbko",[64,68,71],{"title":55,"path":65,"acronym":6,"category":66,"difficulty":47,"description":67},"\u002Fterms\u002Fc\u002Fclaude-code","ai","Claude Code is Anthropic's agentic coding tool that lives in your terminal.",{"title":56,"path":69,"acronym":6,"category":45,"difficulty":47,"description":70},"\u002Fterms\u002Fp\u002Fpermission-mode","Permission mode controls whether your AI agent confirms before acting, auto-approves low-risk moves, or skips all gates entirely.",{"title":54,"path":72,"acronym":6,"category":45,"difficulty":47,"description":73},"\u002Fterms\u002Fv\u002Fvibe-coding","Vibe coding is writing software by vibes — you describe what you want to an AI, it writes the code, and you just... vibe.",1775560876979]