About MoldSim MCP Server
MoldSim MCP Server gives AI assistants injection-molding simulation knowledge, for engineers working with materials, processes, and part design. It combines material data with validation and specification and DFM generation tools.
Key features
- 01Injection-molding simulation Q&A
- 02Properties for 21 materials
- 03Process parameter validation
- 04Simulation specification generation
- 05Material comparison
- 06DFM checklist generation
Use cases
- 01Investigating molding defects and process optimization
- 02Checking parameters against material processing windows
- 03Drafting a simulation plan from a part description
- 04Reviewing part parameters against DFM rules
MCP support
An MCP server that exposes injection-molding knowledge and analysis tools to MCP-compatible AI assistants.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
Tools
6Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| context | string | No | Additional context (material, part type, defect observed) |
| question | string | Yes | Your question about injection molding simulation |
Details
From the project READMEMoldSim MCP Server
Give your AI assistant injection molding expertise.
An MCP (Model Context Protocol) server that gives AI assistants expert knowledge about injection molding simulation. Material properties, process validation, troubleshooting, and simulation specification generation — all available as tools.
Website: moldsim.com
Why
Setting up injection molding simulations (Moldflow, Moldex3D, Cadmould) requires deep knowledge: Cross-WLF viscosity models, Tait PVT data, processing windows, mesh strategies, DFM rules. This MCP server makes that knowledge available to any AI assistant that supports MCP.
Quick Start
Claude Code (recommended)
claude mcp add moldsim -- npx -y moldsim-mcp
That's it. Ask Claude about injection molding — it will automatically use MoldSim tools.
claude "What causes warpage in glass-filled nylon parts?"
claude "Give me the Cross-WLF parameters for PC/ABS"
claude "Validate: ABS at 245C melt, 60C mold, 2.5mm wall"
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%/Claude/claude_desktop_config.json (Windows):
{
"mcpServers": {
"moldsim": {
"command": "npx",
"args": ["-y", "moldsim-mcp"]
}
}
}
From source
git clone https://github.com/kobzevvv/moldsim-mcp.git
cd moldsim-mcp
npm install
npm run build
Then point your MCP client to node dist/index.js.
Tools
query_simulation_knowledge
Free-text Q&A against the knowledge base. Ask about defects, DFM rules, mesh strategies, process optimization.
"Why is my part warping?" → causes, solutions, simulation checks
"What mesh type should I use?" → midplane vs dual-domain vs 3D guidance
"How to reduce cycle time?" → cooling optimization strategies
get_material_properties
Material property lookup. Returns Cross-WLF viscosity, Tait PVT, thermal, processing window, and mechanical data.
21 materials included: ABS, PP Homo, PP Copo, PA6, PA66, PA66-GF30, PC, PC/ABS, POM, HDPE, LDPE, PMMA, PBT, PBT-GF30, PET, PS, HIPS, TPU, SAN, ASA, PPE/PS
"PA66-GF30" → Cross-WLF coefficients, Tait PVT, processing window 275-310°C, fiber effects
"PC" → all properties or filter by group (viscosity, thermal, processing)
validate_process_parameters
Checks your process parameters against the material's processing window. Flags errors, warnings, and provides suggestions.
material: "ABS", melt_temp_C: 280 → ERROR: exceeds max 260°C, risk of degradation
material: "PA66", mold_temp_C: 40 → WARNING: below min 70°C, poor crystallization
generate_simulation_spec
Natural language → structured simulation specification. Describe the part and get back analysis types, mesh recommendations, process conditions, and expected outputs.
"Automotive dashboard panel in PC/ABS, 2.5mm wall, warpage analysis"
→ Fill+Pack+Cool+Warp sequence, dual-domain mesh, process conditions, expected outputs
compare_materials
Side-by-side comparison of 2-4 materials. Processing windows, thermal properties, mechanical data, and auto-generated key differences.
["abs-generic", "pc", "pa66-gf30"] → comparison table with stiffness, shrinkage, heat resistance diffs
generate_dfm_checklist
Design for Manufacturability checklist with pass/warn/fail ratings. Input part parameters and get 15+ DFM rule checks.
material: "ABS", wall: 2mm, rib: 1.5mm, draft: 0.5°
→ 16 checks: wall OK, rib WARN (75% ratio), draft WARN (below 1°), sink risk, venting...
Examples
Ask your AI assistant:
- "What are the Cross-WLF parameters for PA66-GF30?"
- "I'm getting sink marks on my PC housing — what should I check?"
- "Validate my process: ABS at 250°C melt, 70°C mold, 2mm wall"
- "Generate a simulation spec for a PP container with 1.5mm walls"
- "Compare ABS vs PC vs PA66-GF30 for an automotive connector"
- "Run a DFM check: 2mm wall, 1.5mm ribs, 0.5° draft in PC"
- "What mesh type should I use for a thick-walled POM gear (5mm)?"
- "How do I fix warpage in a glass-filled PA66 connector?"
Knowledge Coverage
| Area | Coverage |
|---|---|
| Materials | 21 grades with Cross-WLF, Tait PVT, thermal, mechanical, processing + comparison |
| Defects | Short shot, flash, sink marks, warpage, weld lines, burn marks, jetting, splay, voids, flow marks |
| Simulation | Mesh sensitivity, race tracking, cooling, overpacking, shear rate |
| Process | Cycle time, gate location, V/P switchover, degradation, crystallization, fiber orientation |
| DFM | Wall thickness, ribs, draft, radii, bosses, gating, tolerances, shrinkage, venting |
| Meshing | Midplane, dual-domain, 3D — when to use, element sizing, quality metrics |
Compatible Software
MoldSim MCP provides software-agnostic knowledge that applies to all major injection molding simulation packages:
- Autodesk Moldflow — Adviser & Insight
- Moldex3D — all editions
- CADMOULD — Simcon
- SIGMASOFT — Virtual Molding
- SolidWorks Plastics
Tech Stack
- Runtime: Node.js, TypeScript
- Search: Qdrant vector database (local, embedded)
- Embeddings: HuggingFace transformers
- Protocol: Model Context Protocol (MCP)
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.