reinstate-ai-searchlisted

# Reinstate AI Search ## Principles - Normal SEO is the foundation. - Google does not require special AI markup or AI text files. - `llms.txt` is optional navigation, not a ranking guarantee. - `OAI-SearchBot` controls ChatGPT Search discovery separately from GPTBot training policy. - `PerplexityBot` is the documented Perplexity search crawler. - `Perplexity-User` is a separate user-triggered fetcher, not the search crawler; it can ignore robots policy and needs an explicit CDN/WAF decision. - User-agent strings are spoofable. Where crawler identity matters at the WAF, verify requests against the provider's current published IP ranges. - Crawler access must also work through CDN and WAF layers. - Accessible semantic HTML and accurate ARIA improve both human access and user-triggered agent/browser navigation. - Citation likelihood comes from relevance, evidence, clarity, and authority, not keyword stuffing. ## Workflow 1. Load `reinstate-product-truth`. 2. verify crawler policy. 3. test homepage and key pages with relevant user agents. 4. inspect server logs for 403, 429, and 5xx. 5. audit entity consistency across: - website - README - GitHub About - releases - package metadata - social profiles 6. audit citation units: - definitions - facts - procedures - tables - benchmarks 7. verify visible dates and versions. 8. identify missing primary-source evidence. 9. identify third-party corroboration opportunities. 10. run the fixed AI-que