CAD-Query MCP Server
A server for generating and verifying CAD models using the CAD-Query Python library.
CAD-Query MCP Server
A Model Context Protocol (MCP) server that provides CAD generation and verification tools for Claude Code. This server enables conversational 3D modeling by exposing CAD-Query functionality through MCP tools.
Features
verify_cad_query- Validates CAD-Query generated models against criteriagenerate_cad_query- (Stub implementation) Generates CAD-Query Python scripts from descriptions- CAD-Query Integration - Full CAD-Query support for parametric 3D modeling
- STL/STEP Export - Direct export to 3D printing and CAD formats
- Visual Feedback - SVG generation for model inspection
Installation
# Install dependencies
uv sync
# For development with CAD verification capabilities
uv sync --extra cad
# Test the server
uv run python tests/test_server.py
# Run with MCP Inspector (interactive testing)
uv run mcp dev server.py
Claude Desktop Configuration
Add this to your Claude Desktop configuration file:
macOS
Location: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows
Location: %APPDATA%/Claude/claude_desktop_config.json
Configuration
{
"mcpServers": {
"cadquery-server": {
"command": "python",
"args": ["/path/to/cadquery-mcp-server/server.py"],
"env": {}
}
}
}
MCP Tools
verify_cad_query
Validates a CAD-Query generated model against specified criteria.
Parameters:
file_path(string): Path to the CAD-Query Python fileverification_criteria(string): Description of what to verify
Example:
{
"file_path": "models/coffee_mug.py",
"verification_criteria": "coffee mug with handle, 10cm height, 8cm diameter"
}
Returns:
{
"status": "PASS" | "FAIL",
"message": "Description of result",
"file_path": "Path to verified file",
"criteria": "Verification criteria used",
"details": "Additional verification details"
}
generate_cad_query (Stub Implementation)
Generates CAD-Query Python scripts from natural language descriptions.
NOTE: Currently returns a stub response indicating the feature is not yet implemented.
Parameters:
description(string): Natural language description of the desired 3D modelparameters(string, optional): Specific dimensions or constraints
Example:
{
"description": "Create a coffee mug with a handle, 10cm tall and 8cm diameter",
"parameters": "height=100mm, diameter=80mm, handle_width=15mm"
}
Returns:
{
"status": "NOT_IMPLEMENTED",
"message": "CAD code generation is not yet implemented",
"description": "Input description",
"parameters": "Input parameters",
"details": "Additional information"
}
CAD-Query Script Requirements
All CAD-Query scripts must end with show_object(result):
import cadquery as cq
result = cq.Workplane("XY").box(10, 10, 10)
show_object(result) # Required for processing
Development
Testing
# Test server functionality
uv run python tests/test_server.py
# Interactive testing with MCP Inspector
uv run mcp dev server.py
# Run evaluations
uv run python evaluations/evaluate_verify.py
Extending the Server
The current verify_cad_query implementation is a basic validator. You can enhance it to:
- Parse and validate CAD-Query syntax
- Execute model generation and catch errors
- Analyze resulting geometry dimensions
- Check for specific features and constraints
- Generate detailed validation reports
Dev tools
# formatting
uvx rff format
# running the MCP server
npx @modelcontextprotocol/inspector \
uv \
--directory $(pwd) \
run \
server.py
Servidores relacionados
Scout Monitoring MCP
patrocinadorPut performance and error data directly in the hands of your AI assistant.
Alpha Vantage MCP Server
patrocinadorAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
UUID MCP Provider
Generates timestamp-based unique identifiers using UUID v7.
Code Reasoning
Enhances Claude's ability to solve complex programming tasks through structured, step-by-step thinking.
Remote MCP Server (Authless)
An example of a remote MCP server deployable on Cloudflare Workers without authentication.
Skills-ContextManager
Don’t pollute your AI agent’s context with 1,000 skills. Use Skills-ContextManager, a self-hosted web UI for managing AI skills and workflows by providing skills to an AI agent via MCP only when needed. Simply add skills to your library and enable or disable them with a toggle. Choose whether a skill is always loaded into context or dynamically activated when the AI agent determines it’s needed.
GDB
A GDB/MI protocol server based on the MCP protocol, providing remote application debugging capabilities with AI assistants.
YApi
Interact with the YApi platform using natural language for automated interface management.
OpenRPC MCP Server
Provides JSON-RPC functionality through the OpenRPC specification.
SpecLock
AI constraint engine — persistent memory + active enforcement. Stops AI from breaking locked code. Semantic conflict detection, file-level guards, session continuity. 19 MCP tools.
Sentinel Signal MCP
Agent tools via MCP for workflow scoring, limits/usage, and feedback (trial key supported)
XcodeMCP
An MCP server to control Xcode on macOS using JavaScript for Automation (JXA).