PrusaMCP
An intelligent 3D printing assistant that enables users to analyze meshes, generate optimal print profiles, and automate PrusaSlicer operations through an MCP client. It provides a comprehensive suite of tools for geometric analysis, printability checks, filament recommendations, and post-print diagnostics.
README
PrusaMCP
Intelligent MCP Server for PrusaSlicer — 17 tools to analyze, configure and launch your 3D prints from Claude or any MCP client.
PrusaMCP is not a simple CLI wrapper. It's a 3D printing assistant that analyzes your mesh, recommends optimal settings with justification, diagnoses defects, and can drive PrusaSlicer automatically.
Built with Claude Code.
Features
- Native mesh analysis — Binary/ASCII STL and 3MF parser, zero external dependencies for parsing
- Recommendation engine — Bible FDM + official Prusa knowledge base, justification for every setting
- MVS-based speed capping — Automatic speed calculation based on material Max Volumetric Speed
- 17 MCP tools covering the full print workflow
- Post-print diagnostics — Bible FDM diagnostic tree for 9 common defects
- Filament database — 30+ filaments with temperatures, MVS, recommended surface
Tools
Analysis & Recommendation (no PrusaSlicer needed)
| Tool | Description |
|---|---|
analyze_mesh |
Geometric analysis of STL/3MF (dimensions, volume, overhangs, manifold) |
check_printability |
Issue detection: thin walls, bridges, overhangs, stability |
suggest_orientation |
Test 6 orientations with scoring (overhangs, height, adhesion) |
recommend_profile |
Full profile recommendation with per-setting justification |
generate_prusaslicer_config |
Generate PrusaSlicer-compatible .ini file |
estimate_cost |
Cost estimate (filament + electricity) and print time |
search_filament |
Search filament database (30+ entries) |
print_wizard |
All-in-one: analysis + printability + orientation + profile + questions |
diagnose_print |
Post-print diagnosis (warping, stringing, under-extrusion...) |
PrusaSlicer Integration
| Tool | Description |
|---|---|
slice_prusaslicer |
Run slicing via CLI with G-code stats |
get_current_model |
Get the model currently open in PrusaSlicer |
screenshot_prusaslicer |
Capture PrusaSlicer window (even in background) |
postprocess_gcode |
G-code post-processing (pause, filament change, etc.) |
upload_print |
Upload to OctoPrint / Prusa Connect |
Feedback & Community
| Tool | Description |
|---|---|
submit_feedback |
Submit print feedback after printing |
feedback_stats |
Feedback statistics by material/goal |
export_feedback |
Anonymized community data export |
Installation
git clone https://github.com/Noosbai/PrusaMCP.git
cd PrusaMCP
npm install
npm run build
Configuration
Claude Desktop / Claude Code
Add to your MCP configuration:
{
"mcpServers": {
"prusa-mcp": {
"command": "node",
"args": ["/path/to/PrusaMCP/build/index.js"],
"env": {
"PRUSASLICER_PATH": "C:/Program Files/Prusa3D/PrusaSlicer/prusa-slicer-console.exe"
}
}
}
}
Environment Variables
| Variable | Description | Default |
|---|---|---|
PRUSASLICER_PATH |
Path to prusa-slicer-console.exe |
Auto-detected |
PRUSASLICER_PROFILES_DIR |
PrusaSlicer profiles folder | %APPDATA%/PrusaSlicer |
OCTOPRINT_URL |
Your OctoPrint instance URL | — |
OCTOPRINT_API_KEY |
OctoPrint API key | — |
Usage
Typical Workflow
- Analyze a model:
analyze_meshto get dimensions, overhangs, etc. - Check printability:
check_printabilityto detect potential issues - Get a recommendation:
recommend_profilewith your goal (draft, standard, quality, strong, speed) - Generate config:
generate_prusaslicer_configfor a ready-to-use .ini - Slice:
slice_prusaslicerto generate G-code
Or in one command
print_wizard does everything at once: analysis + printability + orientation + profile + cost estimate + follow-up questions to refine.
Available Goals
| Goal | Description |
|---|---|
draft / prototype |
Fast, quality secondary |
standard |
Good quality/time balance |
quality / fine |
Smooth surface, fine details |
strong / structural |
Maximum mechanical strength |
vase |
Spiral vase mode |
speed |
Everything maxed out |
Supported Materials
PLA, PETG, ABS, ASA, TPU, Nylon, PC — with official Prusa temperatures, Bible FDM MVS values, drying/enclosure warnings.
Knowledge Base
The recommendation engine is built on:
- Bible de l'impression 3D FDM — Comprehensive FDM settings guide for PrusaSlicer
- Official Prusa documentation — Temperatures, MVS, reference profiles
- 9-step calibration workflow (mechanical → PID → Z → EM → MVS → retraction → cooling → PA → accel)
- Diagnostic tree for 9 common defects with causes and fixes
MVS (Max Volumetric Speed) — The Central Parameter
| Material | MVS (mm³/s) |
|---|---|
| PLA | ~15 |
| ABS/ASA | ~11 |
| PETG | ~8 |
| Nylon/PC | ~8 |
| PVA/BVOH | ~4 |
| TPU | ~1-2.5 |
Architecture
src/
├── index.ts # MCP entry point + STDIO transport
├── types.ts # TypeScript interfaces
├── config.ts # PrusaSlicer detection
├── stl-parser.ts # Native binary/ASCII STL parser
├── threemf-parser.ts # 3MF parser (ZIP + XML)
├── mesh-analyzer.ts # Geometric analysis
├── profile-engine.ts # Recommendation engine (Bible FDM)
├── ini-writer.ts # PrusaSlicer .ini generation
├── print-issues.ts # Issue detection + Bible FDM diagnostics
├── orientation.ts # Optimal orientation suggestion
├── cost-estimator.ts # Cost/time estimation
├── filament-db.ts # Filament database
├── prusa-cli.ts # PrusaSlicer CLI wrapper
├── community-data.json # Community data (MVS, diagnostics, best practices)
└── tools/ # 17 MCP tools
├── analyze-mesh.ts
├── check-printability.ts
├── suggest-orientation.ts
├── recommend-profile.ts
├── generate-config.ts
├── estimate-cost.ts
├── search-filament.ts
├── print-wizard.ts
├── diagnose-print.ts
├── slice.ts
├── get-current-model.ts
├── screenshot-prusaslicer.ts
├── postprocess-gcode.ts
├── upload-print.ts
└── feedback.ts
Requirements
- Node.js >= 18
- PrusaSlicer (optional — only needed for slice, screenshot, get_current_model)
- Windows for screenshot feature (uses PrintWindow API)
License
MIT
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。