About lego-oracle
lego-oracle is an MCP server for querying LEGO sets, parts, minifigures, and inventories. It serves builders and collectors with Rebrickable catalog data bundled locally, enabling queries without runtime network calls.
Key features
- 01Search sets by name, theme, year, or piece count
- 02Compare set details and shared parts
- 03Search parts by category, colour, or material
- 04Look up minifigures and their sets
- 05Browse themes and find alternate builds
Use cases
- 01Find sets matching a theme or piece count
- 02Check which sets contain a particular brick
- 03Compare minifigures across sets
- 04Explore alternate builds using a set’s parts
MCP support
Exposes 10 LEGO catalog tools to MCP-compatible clients, including search, lookup, comparison, and discovery tools.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project README
LEGO Oracle
LEGO sets, parts, minifigs, and inventories. An MCP server.
Ask your AI assistant about LEGO sets, find specific bricks, look up minifigures, browse themes, and compare sets. All backed by Rebrickable's catalog (26k sets, 62k parts, 17k minifigs), not hallucinations.
10 tools. Zero config. Works with every MCP-compatible IDE.
Install
npx -y lego-oracle
Add to your IDE
Claude Code
claude mcp add lego-oracle -- npx -y lego-oracle
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%/Claude/claude_desktop_config.json (Windows):
{
"mcpServers": {
"lego-oracle": {
"command": "npx",
"args": ["-y", "lego-oracle"]
}
}
}
Cursor
Add to .cursor/mcp.json:
{
"mcpServers": {
"lego-oracle": {
"command": "npx",
"args": ["-y", "lego-oracle"]
}
}
}
VS Code (Copilot)
Add to .vscode/mcp.json:
{
"servers": {
"lego-oracle": {
"command": "npx",
"args": ["-y", "lego-oracle"]
}
}
}
Windsurf
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"lego-oracle": {
"command": "npx",
"args": ["-y", "lego-oracle"]
}
}
}
Zed
Add to settings.json:
{
"context_servers": {
"lego-oracle": {
"command": {
"path": "npx",
"args": ["-y", "lego-oracle"]
}
}
}
}
Tools
Set Tools
| Tool | Description |
|---|---|
search_sets | Search for sets by name, theme, year, or piece count. Full-text search with filters. |
get_set | Get complete set details: inventory (grouped by part category), minifigures, theme hierarchy. |
compare_sets | Compare 2-4 sets side by side: piece count, year, theme, minifigs, shared parts. |
Part Tools
| Tool | Description |
|---|---|
search_parts | Search parts by name, category, colour, or material. |
get_part | Get part details including available colours and mold/print variants. |
find_part_in_sets | Find which sets contain a specific part (optionally in a specific colour), sorted by quantity. |
Minifig Tools
| Tool | Description |
|---|---|
search_minifigs | Search minifigures by name. |
get_minifig | Get minifig details and every set it appears in. |
Discovery Tools
| Tool | Description |
|---|---|
browse_themes | Browse the LEGO theme hierarchy. Top-level themes or drill into sub-themes with set counts. |
find_mocs | Find community alternate builds (MOCs) for a set's parts. |
Data
All data is embedded at build time from Rebrickable. No network calls at runtime.
| Category | Count |
|---|---|
| Sets | 26,339 |
| Parts | 61,702 |
| Colours | 275 |
| Minifigures | 16,646 |
| Inventory entries | 1,483,652 |
| Part relationships | 35,757 |
| Themes | 490 |
Data is updated daily via GitHub Actions. New npm versions are published automatically when Rebrickable data changes.
Development
npm install
npm run build
npm test # 141 tests
To refresh data from Rebrickable:
npm run fetch-data
npm run build
Attribution
Data from Rebrickable. LEGO is a trademark of the LEGO Group. This project is not produced by or endorsed by the LEGO Group.
License
MIT
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.