Fallback documentation files
If your AI editor cannot use MCP servers or agent skills, use these static documentation files as context for storefront APIs, drop-ins, and Edge Delivery Services.
Available files
Section titled “Available files”These plain-text files are generated from the published documentation each time the site is updated, so AI tools always have current content. Start with llms.txt, which indexes the available documentation. Topic files under _llms-txt/ cover individual subject areas, letting your agent load only the documentation it needs instead of the entire site.
| File | Purpose |
|---|---|
llms.txt | Index of all documentation files, with an overview and links to each one. |
llms-full.txt | Complete documentation in a single file, concatenated in sidebar order. |
llms-small.txt | Complete documentation without the release changelog. Use this file when your AI tool has a limited context window. |
_llms-txt/*.txt | Documentation for individual subject areas, such as drop-ins reference, tutorials, and blocks. llms.txt links to every topic file. |
Prerequisites
Section titled “Prerequisites”Before adding context files, make sure you have the following:
- An AI editor or coding agent that supports project context files, documentation URLs, or pasted context, such as Cursor, Claude Code, or GitHub Copilot Chat.
- A storefront project based on the Adobe Commerce boilerplate or configured with Commerce blocks and drop-ins, and open in your AI editor or coding agent.
- An
AGENTS.md,CLAUDE.md, or equivalent context file in the project root (for agent-based tools).
Add context files
Section titled “Add context files”For agent-based tools, add the llms.txt URL to a project context file so your agent can fetch the documentation index when storefront documentation is needed. For other tools, paste one of these file URLs—or its contents—directly into the chat.
Add this section to your project’s context instructions—for example, AGENTS.md in the project root or a rules file under .cursor/rules/:
## Adobe Commerce storefront documentation
See https://experienceleague.adobe.com/developer/commerce/storefront/llms.txt for Adobe Commerce Storefront documentation.Add this line to CLAUDE.md in your project root:
See https://experienceleague.adobe.com/developer/commerce/storefront/llms.txt for Adobe Commerce storefront documentation.Tools that implement the llms.txt specification automatically follow the index when you provide this URL:
https://experienceleague.adobe.com/developer/commerce/storefront/llms.txtIf your AI tool doesn’t support URL-based context, paste the contents of llms-small.txt or a relevant _llms-txt/*.txt topic file directly into the chat or system prompt.
Example prompt
Section titled “Example prompt”After adding the documentation context, verify the setup by asking a question the documentation answers, such as:
Using the Storefront documentation, how do I display a **Delivery estimate** message in the Product Details drop-in below the existing **Add to Cart** button?A successful setup returns a response that references storefront topics, drop-in APIs, or documentation paths relevant to your project. Adapt the example to your project’s drop-in names and UI labels, then verify the implementation against the live documentation.
Keep context focused
Section titled “Keep context focused”- These files can be large. When possible, give your AI editor or coding agent a specific documentation URL instead so it can fetch the live page, including recent updates and diagrams that are not included in the static files. Use the static files only when your editor or coding agent can’t fetch URLs.
- Paths and examples in this documentation follow the folder layout of the Adobe Commerce boilerplate . Keep your storefront project open so suggested paths match your workspace. If your repository uses a different layout—for example, if you merged the boilerplate into an existing site—tell your agent.
- To focus a single conversation on one topic, paste the corresponding documentation page URL into the chat. Your context file remains available, and the URL narrows only the current conversation.