[{"data":1,"prerenderedAt":113},["ShallowReactive",2],{"blog-post-seo-for-llms":3},[4,16,28,34,44,57,74,82,89],{"component":5,"data":6},"BlogPostHeader",{"slug":7,"kind":8,"title":9,"publishedAt":10,"readMinutes":11,"author":12},"seo-for-llms","article","SEO for LLMs in the AI era","[12 Sep 2026]",8,{"name":13,"role":14,"initials":15},"Michael","Fullstack developer","SW",{"component":17,"data":18},"BlogPostCover",{"cover":19},{"type":20,"lines":21},"text",[22,25],{"text":23,"muted":24,"strike":24},"robots.txt",true,{"text":26,"indent":27},"llms.txt",1.2,{"component":29,"data":30},"BlogPostSource",{"kind":8,"source":31},{"name":32,"url":33},"habr.com","https:\u002F\u002Fhabr.com\u002F",{"component":35,"data":36},"BlogPostChapter",{"id":37,"name":38,"hidden":24,"blocks":39},"intro","Introduction",[40],{"component":41,"data":42},"BlogPostText",{"text":43},"People increasingly ask AI assistants for answers. Those responses are synthesized from web sources that model providers crawl, index and score for reliability. [Continue the intro: why it matters for a small business site.]",{"component":35,"data":45},{"id":46,"name":47,"blocks":48},"what","What is llms.txt",[49,52],{"component":41,"data":50},{"text":51},"[A short explanation: a plain-text map of your site for language models, next to robots.txt and sitemap.xml.]",{"component":53,"data":54},"BlogPostCode",{"language":55,"code":56},"markdown","# Sawa Studio\n> Fullstack developer: fast, SEO-friendly web products.\n## Projects\n- [Qwixit](https:\u002F\u002F[site]\u002Fprojects\u002Fqwixit): fashion marketplace\n- [Brendmart](https:\u002F\u002F[site]\u002Fprojects\u002Fbrendmart): online store",{"component":35,"data":58},{"id":59,"name":60,"blocks":61},"setup","How to set it up",[62,69],{"component":63,"data":64},"BlogPostList",{"ordered":24,"items":65},[66,67,68],"[Step one: list the pages that matter.]","[Step two: write one-line descriptions.]","[Step three: publish at \u002Fllms.txt and link it from robots.txt.]",{"component":70,"data":71},"BlogPostCallout",{"variant":72,"text":73},"tip","[A practical tip, e.g. regenerate the file on every deploy with llms-txt-gen.]",{"component":35,"data":75},{"id":76,"name":77,"blocks":78},"results","Results",[79],{"component":41,"data":80},{"text":81},"[What changed after: mentions in AI answers, traffic, leads.]",{"component":83,"data":84},"BlogPostTags",{"tags":85},[86,87,26,88],"SEO","AI","Next.js",{"component":90,"data":91},"BlogPostRelated",{"items":92},[93,104],{"slug":94,"kind":8,"title":95,"cover":96},"headless-or-classic-cms","Headless or classic CMS",{"type":20,"lines":97},[98,100],{"text":99},"{ headless }",{"text":101,"scale":102,"muted":24,"indent":103},"vs \u003Cclassic \u002F>",0.62,2.4,{"slug":105,"kind":106,"title":107,"cover":108},"llms-txt-gen","tool","[llms-txt-gen]",{"type":109,"title":110,"command":111,"output":112},"terminal","github.com\u002F[user]\u002Fllms-txt-gen","npx llms-txt-gen init","✓ llms.txt generated · 42 pages",1792947429273]