Summary
This server gives Claude direct access to STL 3D model manipulation through numpy-stl. It exposes 30+ tools split across reading and writing meshes, applying transformations like translate, rotate, scale, shear, and mirror, generating geometric primitives from cubes to tori, and creating specialized mechanical components like spur gears, crankshafts, and cam lobes. You'd reach for this when you need to programmatically generate or modify 3D printable parts without opening CAD software. The transformation tools support both simple operations and advanced ones like arbitrary axis rotation and circular arrays. All operations work with both ASCII and binary STL formats, making it straightforward to build parametric model generators or batch processing workflows.
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →
Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →
Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →
Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →
Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →
belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →
mcp-stl
MCP server for editing STL 3D model files. Provides tools for parsing, viewing, and manipulating STL files through the Model Context Protocol.
Install
pip install mcp-stl
MCP Server
mcp-name: io.github.daedalus/mcp-stl
Tools
Reading & Writing
- read_stl_file: Read an STL file and return mesh data (vertices, normals, bounding box)
- get_mesh_info: Get summary information about an STL file without loading full geometry
- write_stl: Write mesh data to an STL file (ASCII or binary format)
Transformations
- translate_stl: Translate (move) the mesh by specified X, Y, Z offsets
- rotate_stl: Rotate the mesh around X, Y, or Z axis by a given angle in degrees
- scale_stl: Scale the mesh by specified X, Y, Z factors
- shear_stl: Apply a shear transformation (shift one axis per unit along another)
- mirror_stl: Mirror (reflect) the mesh across the plane perpendicular to a given axis
- rotate_stl_axis: Rotate the mesh around an arbitrary axis vector (Rodrigues' formula)
- combine_stl: Merge multiple STL files into one
- array_linear: Create N copies of the mesh spaced by a fixed X/Y/Z offset
- array_circular: Create N copies of the mesh at equal angular intervals around an axis
Primitives
- create_cube: Create a cube mesh
- create_box: Create a rectangular box (cuboid) with independent width/height/depth
- create_sphere: Create a sphere mesh
- create_cylinder: Create a cylinder mesh
- create_cone: Create a cone mesh
- create_torus: Create a torus (donut) mesh
- create_plane: Create a flat plane mesh
- create_capsule: Create a capsule (cylinder with hemispherical end caps)
- create_ellipsoid: Create an ellipsoid with independent radii on each axis
- create_frustum: Create a frustum (truncated cone)
- create_tube: Create a hollow cylinder (tube/pipe)
- create_pyramid: Create a regular n-sided pyramid
- create_prism: Create a regular n-sided prism (e.g. hexagonal prism)
- create_hemisphere: Create a hemisphere (dome with flat circular base)
- create_wedge: Create a right-triangular wedge (triangular prism)
Engine-Design Shapes
- create_gear: Create a spur gear mesh (configurable module, teeth, pressure angle)
- create_spring: Create a helical coil spring
- create_connecting_rod: Create a connecting rod with big/small-end bores
- create_crankshaft: Create an N-throw crankshaft
- create_valve: Create a poppet valve (stem + head disc)
- create_camshaft_lobe: Create an eccentric cam lobe with cosine nose profile
Usage
Python API
from mcp_stl import read_stl_file, create_cube, translate_stl
# Read an existing STL file
mesh = read_stl_file("model.stl")
print(f"Faces: {mesh.face_count}")
# Create a primitive
create_cube("cube.stl", size=2.0)
# Transform a mesh
translate_stl("input.stl", "output.stl", x=10.0, y=0.0, z=5.0)
MCP Server
Configure in your MCP client:
{
"mcpServers": {
"mcp-stl": {
"command": "mcp-stl"
}
}
}
Development
git clone https://github.com/daedalus/mcp-stl.git
cd mcp-stl
pip install -e ".[test]"
# run tests
pytest
# format
ruff format src/ tests/
# lint
ruff check src/ tests/
# type check
mypy src/
Featured
Plug Mailtrap into your AI workflow and let it handle the email.
Connect Mailtrap MCP →Agent, run crypto. Access onchain data & trade routes via 1inch.
Install now →Make coding agent sessions - Searchable, Shareable, Vendor-neutral & Scored.
Try For Free →Your agent targets a perfect 10 Code Health score. Deterministic. Every commit.
Try For Free →Integrate web data into your AI product. One API to scrape website & brand data.
Get API Key Now →belt cli automatically finds the best tools and skills for your agent. image, video, music, tts...
one prompt install →Categories
Registryactive
Packagemcp-stl
TransportSTDIO
UpdatedApr 4, 2026
