CLI Tools & IDE Setup

C Claude Code / Claude Desktop

Anthropic's AI assistant with native MCP support.

Configuration

Add to ~/.claude/claude_desktop_config.json:

{
  "mcpServers": {
    "conjure": {
      "command": "uvx",
      "args": ["conjure-mcp"],
      "env": {
        "CONJURE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Alternative: npx

{
  "mcpServers": {
    "conjure": {
      "command": "npx",
      "args": ["-y", "@anthropic-ai/claude-mcp-proxy", "--", "uvx", "conjure-mcp"],
      "env": {
        "CONJURE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Cu Cursor IDE

AI-first code editor with built-in MCP support.

Project Configuration

Create .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "conjure": {
      "command": "uvx",
      "args": ["conjure-mcp"],
      "env": {
        "CONJURE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Global Configuration

Or add to ~/.cursor/mcp.json for all projects.

W Windsurf IDE

Codeium's AI-powered IDE with MCP support.

Configuration

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "conjure": {
      "command": "uvx",
      "args": ["conjure-mcp"],
      "env": {
        "CONJURE_API_KEY": "your_api_key_here"
      }
    }
  }
}

O OpenAI ChatGPT Desktop

OpenAI's desktop app with MCP support (added March 2025).

Configuration

Add to your ChatGPT desktop MCP configuration:

{
  "mcpServers": {
    "conjure": {
      "command": "uvx",
      "args": ["conjure-mcp"],
      "env": {
        "CONJURE_API_KEY": "your_api_key_here"
      }
    }
  }
}

Note: Check OpenAI's documentation for the exact config file location.

Ag OpenAI Agents SDK (Python)

Build custom AI agents with MCP tool support.

Installation

pip install openai-agents

Usage

from agents import Agent
from agents.mcp import MCPServerStdio

# Configure Conjure MCP server
conjure_server = MCPServerStdio(
    command="uvx",
    args=["conjure-mcp"],
    env={"CONJURE_API_KEY": "your_api_key_here"}
)

# Create an agent with Conjure tools
agent = Agent(
    name="CAD Assistant",
    instructions="You are a CAD design assistant. Use the Conjure tools to create 3D models.",
    mcp_servers=[conjure_server]
)

# Run the agent
async with agent:
    result = await agent.run("Create a box 50x30x20mm with filleted edges")
    print(result)