Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions astro.sidebar.mjs
Original file line number Diff line number Diff line change
Expand Up @@ -31,6 +31,7 @@ export function generateSidebar() {
{ label: 'Browser compatibility', link: '/get-started/browser-compatibility/' },
{ label: 'Performance best practices', link: '/get-started/performance/' },
{ label: 'Documentation bundles for AI tools', link: '/get-started/build-with-ai/' },
{ label: 'Dropins MCP', link: '/get-started/dropins-mcp/' },
],
},
{
Expand Down Expand Up @@ -197,6 +198,7 @@ export function generateSidebar() {

// ---------- REFERENCE ----------
{ label: 'Reference', link: '/reference/', attrs: { class: 'sidebar-section-label large' } },
{ label: 'Dropins MCP', link: '/reference/dropins-mcp/' },

// ---------- B2C DROP-INS ----------
{
Expand Down
133 changes: 133 additions & 0 deletions src/content/docs/get-started/dropins-mcp.mdx
Original file line number Diff line number Diff line change
@@ -0,0 +1,133 @@
---
title: Dropins MCP
description: Install and configure the Dropins MCP server to give your AI editor live knowledge of Adobe Commerce drop-in APIs, slots, events, and design tokens.
---

import Link from '@components/Link.astro';
import TableWrapper from '@components/TableWrapper.astro';
import { Tabs, TabItem, Steps, Aside } from '@astrojs/starlight/components';

The Dropins MCP is a [Model Context Protocol](https://modelcontextprotocol.io) server that gives AI editors and agents live, accurate knowledge of Adobe Commerce drop-in components. Instead of relying on training data that may not reflect the latest APIs, the MCP queries a registry generated directly from drop-in source code so slots, events, containers, and API functions are always current.

Once configured, your AI editor can answer questions like _"What slots does CartSummaryList have?"_, scaffold a new block with the right imports, check whether your project references any renamed or removed APIs, and generate working event handler and slot code without guessing.

## Prerequisites

- Node.js 18 or later (`node -v` to confirm)
- An AI editor that supports MCP servers: [Cursor](https://cursor.sh), [VS Code with GitHub Copilot](https://code.visualstudio.com), or [Claude Desktop](https://claude.ai/download)

## Installation

<Steps>

1. **Add the server to your editor configuration.** Choose your editor below and paste the snippet into the correct config file.

<Tabs>
<TabItem label="Cursor">
Add to `~/.cursor/mcp.json` for all projects, or `.cursor/mcp.json` inside a specific project:

```json
{
"mcpServers": {
"dropins": {
"command": "npx",
"args": ["@dropins/mcp"]
}
}
}
```
</TabItem>
<TabItem label="VS Code">
Add to `.vscode/mcp.json` inside your project:

```json
{
"servers": {
"dropins": {
"type": "stdio",
"command": "npx",
"args": ["@dropins/mcp"]
}
}
}
```
</TabItem>
<TabItem label="Claude Desktop">
Add to `claude_desktop_config.json`:

```json
{
"mcpServers": {
"dropins": {
"command": "dropins-mcp"
}
}
}
```

Claude Desktop spawns processes without the shell's PATH, so `npx` may not resolve. Using the installed binary directly (`dropins-mcp`) is more reliable. First install the package globally:

```bash
npm install -g @dropins/mcp
```

If you use `nvm`, replace `dropins-mcp` with the full path to the binary:

```
~/.nvm/versions/node/<version>/bin/dropins-mcp
```
</TabItem>
</Tabs>

2. **Restart your editor.** The Dropins MCP appears in the tool list with 20 tools available.

3. **Verify the connection.** Ask your AI assistant:

> _"Use the dropins MCP — what slots does the CartSummaryList container have?"_

If it returns a list of slot names with context types, the MCP is connected and working.

If the server does not appear, open your editor's MCP logs and confirm that `npx @dropins/mcp` runs without errors in a terminal outside the editor.

</Steps>

## Optional: enable AI-powered documentation search

The `search_commerce_docs` tool provides semantic search across the full Adobe Commerce documentation using the Adobe I/O CLI. Without it, the MCP still works fully local documentation snapshots are always available through `search_docs`.

To enable it:

```bash
npm install -g @adobe/aio-cli
aio auth login
```

## What you can ask

The MCP works best with concrete, task-oriented prompts. A few examples:

- _"What events does the checkout dropin emit when a payment method is selected?"_
- _"Scaffold a block that combines CartSummaryList and OrderConfirmation"_
- _"Are there any stale containers or removed APIs in my commerce-cart block?"_
- _"What GraphQL mutation does checkout use to place an order?"_
- _"Generate a slot implementation that adds a loyalty points badge next to the cart line item price"_
- _"Am I using the latest drop-in versions?"_
- _"What CSS design tokens control button colors?"_

<Aside type="tip">
If your editor does not automatically route questions to the MCP, add _"Use the dropins MCP"_ at the start of your prompt. In Cursor, you can also create a rule in `.cursor/rules/` that instructs the agent to always consult the MCP for drop-in questions.
</Aside>

## Drop-in coverage

The MCP covers the full drop-in suite and SDK packages.

**B2C drop-ins:** Cart, Checkout, Order, User Auth, User Account, Product Details, Product Discovery, Wishlist, Recommendations, Payment Services, Personalization.

**B2B drop-ins:** Quote Management, Requisition List, Company Management, Company Switcher, Purchase Order, Quick Order.

**SDK packages:** Elsie UI components, Event Bus, Fetch GraphQL, Build Tools, reCAPTCHA, Storefront Design tokens.

For the full tools reference all 20 tools with parameters and usage examples see [Dropins MCP reference](/reference/dropins-mcp/).

For troubleshooting connection and configuration issues, see [FAQ](/troubleshooting/faq/#dropins-mcp).
Loading
Loading