Prepare the server locally
Run this once before adding it to Claude Code.
npm install
npm run buildRegister it in Claude Code
claude mcp add brickognize -- node /absolute/path/to/brickognize-mcp/dist/index.jsReplace any placeholder paths in the command with the real path on your machine.
Make your agent remember this setup
brickognize's config, env vars, and the gotchas you hit — recalled in every future Claude Code, Cursor, and Codex session.
npx conare@latestFree · one command · indexes the sessions already on disk. Set up in the browser instead →
What it does
- Identify LEGO parts, sets, and minifigures from local image files
- Support for JPEG, PNG, and WebP image formats
- Batch processing for identifying multiple items in a single call
- Optional inclusion of raw API response data
- Health check tool for monitoring API availability
Tools 6
brickognize_healthCheck API statusbrickognize_identifyIdentify any LEGO item from an imagebrickognize_identify_partIdentify a specific LEGO partbrickognize_identify_setIdentify a LEGO setbrickognize_identify_figIdentify a LEGO minifigurebrickognize_batch_identifyIdentify multiple LEGO items from images in parallelTry it
Original README from NazarLysyi/brickognize-mcp
Brickognize MCP Server
MCP server for identifying LEGO parts, sets, and minifigures from images using the Brickognize API.
Huge thanks to Piotr Rybak for creating the Brickognize service and making LEGO recognition accessible to everyone!
Setup
npm install
npm run build
Configuration
Add to your MCP client config:
{
"mcpServers": {
"brickognize": {
"command": "node",
"args": ["/absolute/path/to/brickognize-mcp/dist/index.js"]
}
}
}
Tools
| Tool | Description |
|---|---|
brickognize_health |
Check API status |
brickognize_identify |
Identify any LEGO item from an image |
brickognize_identify_part |
Identify a specific LEGO part |
brickognize_identify_set |
Identify a LEGO set |
brickognize_identify_fig |
Identify a LEGO minifigure |
brickognize_batch_identify |
Identify multiple LEGO items from images in parallel |
Image Input
All single-image tools accept:
| Parameter | Description |
|---|---|
imagePath |
Absolute path to a local image file (JPEG, PNG, or WebP) |
includeRaw |
Include raw Brickognize API response in output (default: false) |
Batch Tool
brickognize_batch_identify processes multiple images in a single call — significantly faster than calling single-image tools in a loop.
| Parameter | Description |
|---|---|
imagePaths |
Array of absolute paths to local image files (1–20 images) |
type |
"part" | "set" | "fig" | "general" — type of identification (default: "part") |
includeRaw |
Include raw Brickognize API response in each result (default: false) |
Examples
See the examples folder for prompt templates you can use with this MCP server.
Development
npm run dev # Watch mode
npm run build # Compile
npm run lint # ESLint
npm run format # Prettier