BambuStudio MCP Server

A comprehensive MCP (Model Context Protocol) server for Claude macOS app that provides end-to-end 3D printing automation with Bambu Lab printers - from natural language requirements to perfect printed objects.

What This Does

Tell Claude what you need, and it will:

1. Generate a 3D model from your description

2. Scale it to your exact dimensions

3. Slice it with optimized parameters

4. Print it on your Bambu printer

5. Monitor the print via camera

6. Detect any issues and suggest fixes

7. Learn from each print to improve the next one

Example Workflow

You: "I need a tube squeezer for my Gold Bond lotion bottle, it's about 65mm diameter"

Claude:
1. Parses requirements → tube_squeezer, 65mm, heavy_duty
2. Generates parametric model with CadQuery
3. Adjusts for larger size (thicker walls, more infill)
4. Slices with BambuStudio CLI
5. Monitors print via RTSPS camera
6. Records outcome: 88% quality, minor stringing
7. Recommends: +0.5mm retraction, -5°C nozzle temp

Features

Phase 0: Model Generation (NEW!)

• Natural Language Parsing: "I need a squeezer for 65mm bottle" → structured parameters

• Image Analysis: Extract dimensions from photos with rulers

• Parametric Generation: CadQuery/OpenSCAD for precise functional parts

• Template Library: Pre-built customizable designs (squeezers, holders, clips)

• AI Generation: Text-to-3D via Meshy/Tripo3D for organic shapes

Phase 1: Model Preparation

• Model Analysis: Load and analyze STL/3MF files, extract dimensions

• Smart Scaling: Scale based on target dimensions (toothpaste → lotion bottle)

• Parameter Optimization: Adjust slicing for scaled prints

Phase 2: Print Execution & Monitoring

• Printer Control: MQTT connection to Bambu A1/A1 Mini in LAN mode

• Camera Streaming: RTSPS feed at 1 FPS

• Defect Detection: Layer shifts, stringing, warping, spaghetti detection

Phase 3: Iterative Improvement

• Print History: SQLite database of all attempts

• Recommendation Engine: Learn from defects to suggest fixes

• Quality Scoring: 0-100 score based on detected issues

Novice-Friendly Features (NEW!)

• Interactive Wizard: Guided workflow with checkpoints for beginners

• Plain Language Support: "heavy duty" → thick walls, high infill

• Design Review: Catches issues before you print (thin walls, tight clearances)

• Slicing Review: Recommends settings based on use case (functional vs decorative)

• Material Optimization: Auto-adjusts for your specific filament

Supported Materials

Pre-configured profiles for:

• Bambu Basic PLA - General purpose, 220°C nozzle, up to 300mm/s

• Bambu PETG Translucent - Functional parts, 250°C nozzle

• Prusament PC Blend - High strength, 270°C nozzle (keep parts small on A1)

• Generic PETG - Outdoor/water-resistant, 245°C nozzle

• Generic TPU 95A - Flexible, 230°C nozzle, 40mm/s max, NO AMS

Supported Nozzles

All A1/A1 Mini nozzle configurations:

• 0.2mm Stainless Steel - Fine detail, miniatures

• 0.4mm Stainless Steel - Default, best all-around

• 0.4mm Hardened Steel - For carbon fiber/glow filaments

• 0.6mm Hardened Steel - Faster prints, functional parts

• 0.8mm Hardened Steel - Draft prints, large parts

Requirements

• macOS with BambuStudio installed

• Bambu Lab A1/A1 Mini printer in LAN mode

• Python 3.10+

• Printer access code (from printer settings)

Installation

cd bambustudio-mcp

# Basic installation
pip install -e . --break-system-packages

# With CAD generation (recommended)
pip install -e ".[cad]" --break-system-packages

# With everything
pip install -e ".[all]" --break-system-packages

Configuration

This project uses 1Password CLI for secure secret management. Never store API keys or credentials in plain text files.

Setting Up 1Password CLI

1. Install 1Password CLI:

   brew install --cask 1password-cli

2. Sign in and enable shell integration:

   op signin

3. Store your secrets in 1Password. Create items with these fields:

   • Bambu Printer: ip_address, access_code, serial_number

   • Meshy (optional): api_key

   • Tripo3D (optional): api_key

Running with 1Password CLI

Use op run to inject secrets as environment variables:

op run --env-file=.env.template -- python -m bambustudio_mcp

Create a .env.template file (safe to commit - contains only references, not secrets):

# Printer connection (required)
BAMBU_PRINTER_IP=op://Personal/Bambu Printer/ip_address
BAMBU_ACCESS_CODE=op://Personal/Bambu Printer/access_code
BAMBU_SERIAL=op://Personal/Bambu Printer/serial_number

# BambuStudio path (usually auto-detected)
BAMBUSTUDIO_PATH=/Applications/BambuStudio.app/Contents/MacOS/BambuStudio

# Optional: AI 3D generation
MESHY_API_KEY=op://Personal/Meshy/api_key
TRIPO3D_API_KEY=op://Personal/Tripo3D/api_key

Claude Desktop Configuration

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "bambustudio": {
      "command": "op",
      "args": [
        "run",
        "--env-file=/path/to/bambustudio-mcp/.env.template",
        "--",
        "python",
        "-m",
        "bambustudio_mcp"
      ]
    }
  }
}

This ensures secrets are never stored in plain text and are securely injected at runtime.

Available Tools (33 total)

Novice-Friendly Wizard Tools (NEW!)

Model Generation

Model Preparation

Printer Control

Camera & Quality

Iteration Tracking

Architecture

bambustudio_mcp/
├── server.py              # MCP server with all 33 tools
├── config.py              # Environment configuration
├── materials/             # NEW: Filament & nozzle profiles
│   ├── filaments.py       # 5 pre-configured filament profiles
│   └── nozzles.py         # A1 nozzle configurations
├── wizard/                # NEW: Novice-friendly workflow
│   ├── guided_workflow.py # Interactive checkpoints
│   ├── design_review.py   # Design suggestions
│   ├── slicing_review.py  # Quality recommendations
│   ├── novice_parser.py   # Plain language → parameters
│   └── material_optimizer.py # Material-based adjustments
├── generator/             # Model generation
│   ├── requirements.py    # Natural language parsing
│   ├── image_analyzer.py  # Photo dimension extraction
│   ├── parametric.py      # CadQuery/OpenSCAD generation
│   ├── templates.py       # Template library
│   └── ai_generator.py    # Meshy/Tripo3D integration
├── models/
│   ├── analyzer.py        # STL/3MF analysis
│   └── scaler.py          # Dimension scaling
├── slicer/
│   ├── cli.py             # BambuStudio CLI
│   ├── parameters.py      # Slicing parameters
│   └── profiles.py        # Preset management
├── printer/
│   ├── mqtt_client.py     # MQTT communication
│   ├── controller.py      # Job control
│   └── status.py          # Status parsing
├── camera/
│   ├── stream.py          # RTSPS capture
│   └── detector.py        # Defect detection
└── iteration/
    ├── tracker.py         # SQLite history
    └── recommender.py     # Parameter recommendations

Novice-Friendly Usage

For CAD beginners, use the guided workflow:

# Start with your plain language description
workflow = await bambustudio_start_guided_workflow({
    "description": "I need a heavy duty squeezer for my 65mm lotion bottle"
})

# The system parses your description:
# - "heavy duty" → wall_thickness: 3mm, infill: 30%
# - "65mm" → tube_diameter: 65mm
# - "squeezer" → category: tube_squeezer

# Then guides you through checkpoints:
# 1. Requirements confirmation
# 2. Design review (catches thin walls, tight tolerances)
# 3. Material selection (suggests TPU if flexible needed)
# 4. Nozzle selection (recommends 0.4mm for standard parts)
# 5. Quality settings (functional vs decorative)
# 6. Final review before generation

Plain Language Examples

The system understands:

• "heavy duty" → thick walls (3mm), high infill (30%)

• "snug fit" → clearance: 0.2-0.3mm

• "loose fit" → clearance: 0.8-1.0mm

• "flexible" → suggests TPU material

• "waterproof" → suggests PETG

• "heat resistant" → suggests PC or PETG

Examples

See examples/ directory:

• complete_workflow.py - Full end-to-end example

• tube_squeezer_workflow.py - Scaling example

• generation_tools_reference.md - Generation tool docs

• mcp_tools_reference.md - All tools reference

License

MIT