For AI Agents

Everything on this site is machine-readable: Markdown URLs, a full index, a question-answering endpoint, and the Avonni MCP server.

🚧 Early access. Build with AI is rolling out gradually and improving with every release. Something unclear, missing, or broken? Tell us: your feedback directly shapes what we improve next.

This documentation is machine-readable end-to-end. If you are an AI agent, or you build with one, here is everything you can consume programmatically. No scraping needed.

Read any page as Markdown

Append .md to any page URL to get clean Markdown instead of HTML:

https://docs.avonnicomponents.com/experience-cloud/experience-components/ax-map.md

Get the full index

The site publishes a complete index of every page, and a full-corpus export:

https://docs.avonnicomponents.com/llms.txt
https://docs.avonnicomponents.com/llms-full.txt

llms.txt lists every page with its Markdown URL. llms-full.txt returns the entire documentation corpus in one response.

Ask the documentation a question

Any .md URL accepts an ask query parameter and answers in Markdown, with sources. An optional goal parameter tailors the answer to what you are trying to accomplish:

GET https://docs.avonnicomponents.com/experience-cloud/welcome.md?ask=<question>&goal=<end_goal>

Query the component catalog

Prose is one thing; the structured truth is the Avonni MCP server. It documents every component's properties, interactions, and styling hooks, and updates with each release:

https://mcp.avonnicomponents.com

Use the toolset parameter (experience for this package) on list_components and list_interactions, or filter the whole server with ?toolsets=experience. Full setup in Setup.

Install the workflows

The avonni/skills repository ships the agent workflows for creating and updating Avonni artifacts:

npx skills add avonni/skills

Human reading this? All of the above powers the Build with AI experience: connect your assistant once and it uses these sources for you.