PocketBase MCP Server for Cursor AI

Integration of PocketBase with Cursor AI through the Model Context Protocol (MCP). This server allows Cursor AI to directly interact with PocketBase databases, supporting collection management, record operations, and many other functionalities.

Features

$3

- Create and manage collections with custom schemas
- Migrate collection schemas with data preservation capabilities
- Advanced index management (create, delete, list)

$3

- CRUD operations (Create, Read, Update, Delete) for records
- Advanced querying with filtering, sorting, and aggregation
- Bulk import/export capabilities

$3

- Generate PocketBase schemas from TypeScript interfaces
- Generate TypeScript interfaces from PocketBase collections
- Analyze collection data and provide insights

Installation

$3

``bash
npm install -g pocketbase-cursor-mcp
`

or

`bash
pnpm add -g pocketbase-cursor-mcp
`

$3

You can configure the PocketBase MCP Server using environment variables or command line arguments:

#### Using environment variables

Create a .env file in the root directory of your project:

`
POCKETBASE_URL=http://127.0.0.1:8090
POCKETBASE_ADMIN_EMAIL=your-admin@example.com # Optional
POCKETBASE_ADMIN_PASSWORD=your-password # Optional
`

#### Using command line arguments

`bash
pocketbase-cursor-mcp --url=http://127.0.0.1:8090 --admin-email=your-admin@example.com --admin-password=your-password
`

#### Available options

| Command line arg | Environment variable | Description |
|------------------------|-----------------------------|----------------------------------------|
| --url, -u | POCKETBASE_URL | PocketBase server URL (required) |
| --admin-email, -e | POCKETBASE_ADMIN_EMAIL | Admin email (optional) |
| --admin-password, -p | POCKETBASE_ADMIN_PASSWORD| Admin password (optional) |
| --data-dir, -d | POCKETBASE_DATA_DIR | Custom data directory path (optional) |
| --port | PORT | HTTP server port (optional) |
| --host | HOST | HTTP server host (optional) |

Use pocketbase-cursor-mcp --help to view all options.

Cursor AI Configuration

$3

1. Open Cursor AI
2. Open Settings (or press Cmd+, on macOS, Ctrl+, on Windows/Linux)
3. Select the "AI" tab
4. Scroll down to "Model Context Protocol Servers"
5. Add a new configuration with the following information:

Name: pocketbase
Command: npx
Args: pocketbase-cursor-mcp --url=http://127.0.0.1:8090

Or directly provide the path to the executable:

Command: Path to node executable (e.g., /usr/bin/node)
Args: Path to the executable file along with parameters (e.g., /usr/local/bin/pocketbase-cursor-mcp --url=http://127.0.0.1:8090)

Usage in Cursor AI

After configuration, you can use PocketBase MCP in Cursor AI by adding commands like the following to the editing interface:

`
Create a PocketBase collection from the following TypeScript interface:

interface User {
username: string;
email: string;
isActive: boolean;
age?: number;
profile: UserProfile;
}

interface UserProfile {
bio: string;
avatar?: string;
socialLinks: string[];
}
`

or

`
Generate TypeScript interfaces from the collections in my PocketBase database.
`

or

`
Analyze the data in the "products" collection and provide insights.
`

Available Tools

$3


- create_collection: Create a new collection
- create_record: Create a new record
- list_records: List records with optional filters
- update_record: Update an existing record
- delete_record: Delete a record
- get_collection_schema: Get detailed schema for a collection
- ... and many more tools

$3

- generate_pb_schema: Generate PocketBase schema from TypeScript interfaces
- generate_typescript_interfaces: Generate TypeScript interfaces from PocketBase collections
- analyze_collection_data: Analyze data in a collection

Usage Examples

$3

`typescript
const schema = await mcp.use_tool("pocketbase", "generate_pb_schema", {
sourceCode:
interface Product {
name: string;
price: number;
description: string;
isAvailable: boolean;
tags: string[];
}
,
options: {
includeTimestamps: true
}
});

const collection = await mcp.use_tool("pocketbase", "create_collection", {
name: "products",
schema: schema[0].schema
});
`

$3

`typescript
const interfaces = await mcp.use_tool("pocketbase", "generate_typescript_interfaces", {
options: {
includeRelations: true
}
});

// Interfaces can be used in your TypeScript project
`

$3

`typescript
const analysis = await mcp.use_tool("pocketbase", "analyze_collection_data", {
collection: "products",
options: {
sampleSize: 500
}
});

// View insights about your data
console.log(analysis.insights);
``

Contributing

Contributions are always welcome! Please create an issue or pull request.

License

MIT