Audit and improve `/llms.txt`, a curated starting point for AI tools browsing documentation. Check its location, the usefulness and accessibility of linked pages, and its separation from existing crawl and indexing controls.
Use when auditing public documentation portals, API references, help centers, SDK docs, or large knowledge bases. Check the final file served at `/llms.txt` and verify that the linked pages are stable, high-value, and accessible without relying on a JavaScript-only interface.
Review /llms.txt on a public documentation site and organize an entry point that helps AI tools reach important information. The skill checks whether the file is at the root of the production domain and whether it selects useful, stable pages, such as getting-started guides and API references. It also checks whether linked pages return HTTP 200 and whether their content is accessible without relying entirely on JavaScript.
Use it for sites with many public pages, including API references, SDK documentation, and help centers. It helps when you want to select the most worthwhile documents instead of listing every URL, or when you need to review an existing delivery configuration.
/llms.txt and the pages it links to."llms.txt around our getting-started guides and API documentation."llms.txt does not replace robots.txt or an XML sitemap. Avoid linking to private documents or pages that require authentication. A companion llms-full.txt, which collects the full text, is optional. Consider it when you can keep its contents updated over time.
This overview was written by AI (OpenAI Codex) from the public SKILL.md. Check the original below for the exact behavior.
AI rated this skill on four points using only the contents of its SKILL.md. Stars and popularity are not taken into account.
The total counts each point as A = 25, B = 15, C = 5. Rated by AI (OpenAI Codex) as of Oct 23, 2026. Treat it as a guide and read the original before installing.
Review the instructions the agent will receive before you install.
Documentation portals often bury the most important pages behind dense navigation, search UIs, or framework-specific layouts. A curated llms.txt gives AI tools a cleaner entry point without replacing the normal crawl and indexing mechanisms of the web described by the llms.txt convention.
llms.txt at the root (/llms.txt on the production domain) if you run a docs-heavy sitellms-full.txt only as an optional companion when you can keep it currentrobots.txt; llms.txt does not replace themCheck whether this documentation-heavy site publishes llms.txt at the root and whether it contains a curated set of stable, high-value links such as getting-started guides, API references, concepts, and examples. Confirm that linked URLs return HTTP 200 and that llms.txt is not being treated as a replacement for robots.txt or XML sitemaps.
Add llms.txt at the site root with a short description of the project and a curated list of the most useful documentation URLs. If the docs set is large and you can maintain it, add an optional llms-full.txt companion and link to it from llms.txt. Keep crawl directives in robots.txt.
Explain what llms.txt is, why it can help documentation-heavy sites give AI tools a cleaner starting point, why llms-full.txt is optional, and why traditional crawl controls still belong in robots.txt.
Review route handlers, static files, or framework metadata output related to llms.txt. Flag sites that publish the file outside the root path, link to low-value or unstable pages, expose private/auth-only docs, or imply that llms.txt replaces crawl/indexing controls like robots.txt or sitemaps.
For full implementation details, code examples, and framework-specific guidance,
see references/rule.md.
No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Create and review custom 404 pages that help visitors recover from broken or moved links through search and navigation. Check the rendered page and HTTP response so missing content returns a proper 404 status.
Audit or create a dedicated About page that explains a website’s purpose and identifies its people or organization. Review the rendered output and supporting information, including expertise and contact details, to find gaps.
Review and add automated accessibility checks for components, pages, and continuous integration workflows. Use axe-based tools to detect structural and attribute issues, while identifying checks that still need keyboard or screen reader testing.
Review sign-in, registration, multi-factor authentication, and account recovery for accessibility barriers. Find blocked paste, difficult code entry, and inaccessible backup methods, then suggest authentication paths that support password managers, autofill, or device approval.
Review form errors, success messages, and live updates so screen reader users receive important notifications. Check the HTML delivered to the browser, announcement settings, display duration, and dismissal controls, then identify specific elements that need changes.
Review HTML tables so screen reader users can understand how headers and data relate. Check header markup, table descriptions, and merged cells, then identify specific fixes and ways to verify them.