Canonical AI support data

docs/data/ contains small canonical source datasets used to build supplementary bot-friendly cannabis reference files. Keep these files conservative, plain-language, and reviewable.

Generated AI support files are written to docs/public/ai/. Do not manually edit generated support files there when the data comes from docs/data/; update the canonical source file and regenerate instead.

Regenerate the support layer with:

npm run generate:ai-support

This support generation is separate from the approved primary AI dataset files generated by npm run generate:ai-dataset.