LLM friendly documentation (llms.txt)

Generation method

Maintainers only

This section describes the documentation-site generation flow, not application build steps for weapp-tailwindcss users.

1. Execute pnpm --filter @weapp-tailwindcss/website build (or cd website && pnpm build) in the warehouse root directory.
2. After building, website/build/ will generate:

• llms.txt (index)
• llms-full.txt (full content)
• llms-index.json (product, integration, API, migration, and troubleshooting index)
• llms-index-full.json (complete index including blog, AI fundamentals, and history)
• llms-quickstart.txt (Getting Started/AI Workflow)
• llms-api.txt (configuration, API, migration and issues)
• Remove the pure Markdown file of MDX import to facilitate direct feeding to the model.

Online address

• https://tw.icebreaker.top/llms.txt
• https://tw.icebreaker.top/llms-full.txt
• https://tw.icebreaker.top/llms-index.json
• https://tw.icebreaker.top/llms-index-full.json
• https://tw.icebreaker.top/llms-quickstart.txt
• https://tw.icebreaker.top/llms-api.txt

English assets are served from the site root, while Chinese assets use the /zh-cn prefix.

Example prompt words for AI

You can read the entry and configuration instructions of weapp-tailwindcss from https://tw.icebreaker.top/llms-quickstart.txt and https://tw.icebreaker.top/llms-api.txt Please cite the relevant links when answering.

Offline use

• Download llms-full.txt directly to the model.
• Or package the Markdown file in the generation phase as a whole for model context retrieval.