{"Status":true,"Message":"","Response":{"post":{"postuid":"ff874abd-ec5a-43af-b6c5-472bb4997b04","tenantuid":"d8b744fc-2e70-4089-bb80-dd1d08f6c7b2","projectuid":"fc6490ac-7527-4f49-b06e-46f701280e85","title":"Using LLMS.txt","slug":"article/using-llmstxt","html":"\u003Ccite class=\u0022warning\u0022\u003E\u003Cspan class=\u0022title\u0022\u003EWarning\u003C/span\u003E\u003Cp\u003Ellms.txt is a new and emerging standard.\u003C/p\u003E\u003C/cite\u003E\u003Cp\u003EHelpGuides provides native support for the \u003Ccode\u003Ellms.txt\u003C/code\u003E standard, so AI agents, LLMs and developer tools (like Cursor, Perplexity and Claude) can find and understand your documentation efficiently.\u003C/p\u003E\u003Cp\u003EBy providing a machine-readable map of your content, it removes the \u0022noise\u0022 of HTML headers and navigation menus, so AI models can focus on the content they need to give accurate answers.\u003C/p\u003E\u003Cp\u003EAll articles published are automatically \u003Ca href=\u0022/article/viewing-in-markdown\u0022\u003Eavailable in Markdown format\u003C/a\u003E.\u003C/p\u003E\u003Ch2 id=\u0022available_endpoints\u0022\u003EAvailable Endpoints\u003C/h2\u003E\u003Cul\u003E\u003Cli\u003E\u003Cb\u003Ellms.txt\u003C/b\u003E at \u003Ccode\u003Ehttps://YOUR-DOMAIN/llms.txt\u003C/code\u003E: a short index with your site\u0027s name and description, a link to \u003Ccode\u003Ellms-full.txt\u003C/code\u003E, and the articles you\u0027ve chosen to feature, each with its meta description.\u003C/li\u003E\u003Cli\u003E\u003Cb\u003Ellms-full.txt\u003C/b\u003E at \u003Ccode\u003Ehttps://YOUR-DOMAIN/llms-full.txt\u003C/code\u003E: the full text of every published article in one Markdown file, with each article\u0027s URL, last updated date and description. Ideal for giving an LLM your whole library in a single request. It\u0027s refreshed at most every three hours.\u003C/li\u003E\u003C/ul\u003E\u003Cp\u003EA generated \u003Ccode\u003Ellms.txt\u003C/code\u003E looks like this:\u003C/p\u003E\u003Cdiv class=\u0022code_wrapper\u0022\u003E\u003Cdiv class=\u0022code\u0022 data-language=\u0022markdown\u0022\u003E# ACME Documentation\n\n\u0026gt; Help and how-to guides for ACME Cloud\n\nThis file highlights evergreen, structured, and authoritative content suitable for citation.\n\n## Core Content\n\n- [Full Documentation](/llms-full.txt): A single-file version containing the full text of all articles for deep context.\n\n- [Getting Started](https://docs.acme.com/article/getting-started): Set up your ACME Cloud account in five minutes.\u003C/div\u003E\u003C/div\u003E\u003Ch2 id=\u0022choose_the_articles_in_llmstxt\u0022\u003EChoose the articles in llms.txt\u003C/h2\u003E\u003Cp\u003EFeature 3 to 10 of your most important articles. To feature one, open it in the editor and click the star icon above the title; click it again to remove it. Starred articles are included automatically, as long as they\u0027re published.\u003C/p\u003E\u003Cp\u003ETo see which articles are featured, open \u003Cb\u003EAI Settings\u003C/b\u003E and click the \u003Cb\u003Ellms.txt\u003C/b\u003E tab. It lists each featured article and links to your live file. See \u003Ca href=\u0022/article/ai-settings\u0022\u003EAI Settings\u003C/a\u003E.\u003C/p\u003E\u003Ccite class=\u0022recommended\u0022\u003E\u003Cspan class=\u0022title\u0022\u003ERecommended\u003C/span\u003E\u003Cp\u003Ellms.txt is useful for discovery. If you know an LLM or agent will use your HelpGuides content, connect it with Model Context Protocol instead.\u003C/p\u003E\u003C/cite\u003E\u003Cp\u003ESee \u003Ca href=\u0022/article/model-context-protocol\u0022\u003EModel Context Protocol\u003C/a\u003E. Your site\u0027s \u003Ccode\u003Erobots.txt\u003C/code\u003E asks search engine crawlers to skip \u003Ccode\u003Ellms-full.txt\u003C/code\u003E and the Markdown versions of your articles.\u003C/p\u003E\u003Ch2 id=\u0022benefits_for_ai_workflows\u0022\u003EBenefits for AI Workflows\u003C/h2\u003E\u003Cul\u003E\u003Cli\u003E\u003Cb\u003EReduced token usage:\u003C/b\u003E Serving Markdown instead of heavy HTML keeps AI interactions within context limits and makes them faster.\u003C/li\u003E\u003Cli\u003E\u003Cb\u003EImproved accuracy:\u003C/b\u003E Only published articles are included, so AI models don\u0027t learn from drafts or outdated content.\u003C/li\u003E\u003Cli\u003E\u003Cb\u003ETooling support:\u003C/b\u003E Works with the \u003Cb\u003ECursor @Docs\u003C/b\u003E feature. Point Cursor at your documentation URL and it finds the \u003Ccode\u003Ellms.txt\u003C/code\u003E file to build a local index of your guides.\u003C/li\u003E\u003C/ul\u003E\u003Cp\u003ERelated: \u003Ca href=\u0022/article/viewing-in-markdown\u0022\u003EViewing in Markdown\u003C/a\u003E, \u003Ca href=\u0022/article/supporting-ai-crawlers-with-aitxt\u0022\u003ESupporting AI Crawlers with ai.txt\u003C/a\u003E.\u003C/p\u003E","publish_status":0,"post_type":"Article","authoruid":"3dde8c16-763a-4a2b-ae0b-1d8c50c62e3d","author":{"authoruid":"3dde8c16-763a-4a2b-ae0b-1d8c50c62e3d"},"featured_image_updating":false,"meta_description":"LLMS.txt guides AI systems to your site\u0027s best, LLM-friendly content for ingestion, understanding, and citation\u2014like a sitemap for AI tools.","display_toc":true,"has_workingcopy":false,"allow_indexing":true,"total_views":788,"date_published":"2025-06-10T14:13:00","date_updated":"2026-10-04T21:50:20.893","date_created":"2025-06-10T14:12:26.747"}}}