Audit what ChatGPT and Claude say about you. Fix the gaps with an llms.gist file in your repo.
Run your URL through both. See where they're right, where they're wrong, and where they invent things.
Patches what AI misses about your product. Drop it in your repo as llms.gist, or paste into ChatGPT.
Just markdown with a tiny schema. Cursor, Claude Code, ChatGPT — they all read it.
Re-audit when you ship a feature, change positioning, or quarterly. Manual — you decide.