Cursor scores 75/100 (Grade C), passing 12 of 22 checks (55%). The documentation is broadly accessible to AI agents but has 6 areas that could be improved to better serve automated tooling.
# Agent Score Fix Report — Cursor URL: https://cursor.com/docs Score: 75/100 (Grade C) I need help improving the AI-readiness of the documentation at https://cursor.com/docs. Agent Score found 4 failing checks and 2 warnings. ## Failing Checks (4) - [content-discoverability] Llms Txt Valid: llms.txt exists but contains no parseable links - [content-discoverability] Llms Txt Directive: No llms.txt directive found in any of 10 sampled pages - [page-size] Page Size Html: 7 of 10 sampled pages convert to over 100K chars (max 351K, 0% boilerplate) - [page-size] Content Start Position: 10 of 10 sampled pages have content starting past 50% (worst 100%) ## Warnings (2) - [markdown-availability] Content Negotiation: Content negotiation partially supported: 0 correct type, 10 wrong type, 0 HTML only - [page-size] Page Size Markdown: 1 of 10 pages between 50K–100K chars (max 56K) ## Fix Instructions For each issue above, please: 1. Analyze the documentation site at https://cursor.com/docs 2. Implement the specific fix 3. Verify the fix would cause the check to pass ### Common fixes: - **No llms.txt**: Create /llms.txt following https://llmstxt.org — list all doc pages in markdown format - **No .md URL support**: Configure your docs platform to serve pages at equivalent .md URLs (e.g. /docs/quickstart.md) - **No content negotiation**: Return markdown when request includes Accept: text/markdown header - **Large page size**: Reduce nav boilerplate, inline scripts, and repetitive markup - **No sitemap**: Generate /sitemap.xml listing all documentation URLs - **Auth walls**: Ensure docs pages return 200 without requiring login cookies or tokens - **No Last-Modified header**: Configure your server/CDN to include Last-Modified response headers - **Tab content hidden**: Ensure tabbed content is rendered in the HTML (not JS-only) so agents can read all variants ## Run afdocs Locally for More Detail To get deeper visibility into what's failing, run afdocs against your docs: npx afdocs check https://cursor.com/docs --fixes --verbose - **--fixes**: Adds "Fix:" lines to the output for each warn/fail check with actionable remediation steps - **-v, --verbose**: Shows per-page details (specific URLs, character counts, error codes) for checks with issues — useful for per-URL visibility into what's failing
CHECK RESULTS
How your docs scored
llms-txt-validllms.txt exists but contains no parseable links
llms-txt-directiveNo llms.txt directive found in any of 10 sampled pages
llms-txt-existsllms.txt found at 1 location(s)
llms-txt-sizellms.txt is 8,794 characters (under 50,000 threshold)
llms-txt-links-resolveNo HTTP(S) links found in llms.txt
llms-txt-links-markdownNo HTTP(S) links found in llms.txt
page-size-html7 of 10 sampled pages convert to over 100K chars (max 351K, 0% boilerplate)
content-start-position10 of 10 sampled pages have content starting past 50% (worst 100%)
page-size-markdown1 of 10 pages between 50K–100K chars (max 56K)
rendering-strategyAll 10 sampled pages contain server-rendered content
auth-gate-detectionAll 10 sampled pages are publicly accessible
auth-alternative-accessAll docs pages are publicly accessible; no alternative access paths needed
markdown-content-parityAll 10 pages have equivalent markdown and HTML content (avg 0% missing)
cache-header-hygieneAll 11 endpoints have appropriate cache headers
llms-txt-freshnessNo page URLs found in llms.txt
content-negotiationContent negotiation partially supported: 0 correct type, 10 wrong type, 0 HTML only
markdown-url-support10/10 sampled pages support .md URLs (100%)
tabbed-content-serializationNo tabbed content detected across 10 sampled pages
section-header-qualityNo tabbed content found; header quality check not applicable
markdown-code-fence-validityAll 114 code fences properly closed across 11 pages
http-status-codesAll 10 sampled pages return proper error codes for bad URLs
redirect-behaviorNo redirects detected across 10 sampled pages