skppy — SketchUp .skp Toolkit

skppy (reads "skippy") is a Python toolkit for reading SketchUp (.skp) files and writing the confirmed modern-format public model graph. It does not require SketchUp or its C API, and it uses NumPy for vector, matrix, UV, and image-alpha operations.

It provides a Pythonic object model independent of SketchUp's C API. It reads the ZIP/VFF container used by newer files and the earlier CArchive container. Compatibility samples cover the supported save generations and verify format detection and version dispatch; semantic coverage still varies by class and version. skppy.load() detects the container and selects the correct parser automatically.

Continuous integration tests Python 3.11 through 3.14 on Linux, macOS, and Windows. The Blender addon requires Blender 4.2 or newer; its headless integration suite runs against Blender 4.5 LTS and Blender 5.2 LTS.

In addition to the skppy library, this repo contains a self-contained Blender addon for importing and exporting .skp files. The addon converts geometry, reusable components, collection instances, materials/textures, UVs, tags, cameras/scenes, text annotations, and scalar custom properties.

Features

• Read ZIP/VFF .skp files emitted for SketchUp 2021 and later
• Read pre-ZIP CArchive .skp files into the same public model classes
• Parse geometry (vertices, edges, faces, curves, arc curves, guide points/lines, section planes)
• Parse component definitions and instances (including nested groups and images)
• Parse materials with colors, textures, and PBR properties (metallic, roughness)
• Parse layers (tags) and layer folders
• Parse cameras, scenes (saved views), rendering options, shadow info, axes
• Parse edge flags (soft/smooth/hidden bits)
• Parse face UV projections and front/back materials
• Parse line styles, fonts, text styles, dimension styles
• Parse environment data, sun data, options manager, attribute dictionaries
• Triangulate faces with holes, or split single-hole faces into two n-gons
• Builder and writer API for modern models and SketchUp Make 2017 geometry
• Blender addon for importing and exporting .skp files through Blender's File menu

Quick Start

pip install skppy

import skppy

# Load a .skp file
model = skppy.load("my_model.skp")
print(model.header.version_string)       # e.g. "{26.1.103}"
print(len(model.entities.faces))         # number of root-level faces
print(len(model.materials))              # number of materials
print(len(model.definitions))            # number of component definitions

# Traverse geometry
for face in model.entities.faces:
    print(face.id, face.plane, face.front_material_id)
    triangles = face.triangulate(model.entities)

# Inspect materials
for mat in model.materials:
    print(f"{mat.name}: color={mat.color}, metallic={mat.metallic}")

# Look at component definitions
for defn in model.definitions:
    print(f"{defn.name}: {len(defn.entities.faces)} faces")

Creating a model from scratch

import skppy

model = skppy.new_model()
brick = model.add_material("Brick", color=skppy.Color(180, 80, 60))
defn = model.add_definition("Wall")
face = defn.entities.add_face([(0,0,0),(300,0,0),(300,250,0),(0,250,0)])
face.front_material_id = brick.id
model.entities.add_instance(defn, transform=skppy.Transform.identity())
model.save("wall.skp")

Running tests

make test

Run the isolated Blender extension integration with make test-blender.

Release notes are maintained in CHANGELOG.md. Version tags build and test both the Python distribution and bundled Blender extension before publishing their artifacts.

Project

• Author and maintainer: Dante Dex (dante.dex.arch@gmail.com)
• Source: dantedex/skppy
• Documentation: dantedex.github.io/skppy
• Conformance suite: dantedex/skppy-tests

Development and AI Usage

This package used AI assistance during research, implementation, tests, and documentation. All AI-assisted changes were reviewed and edited by a human developer, who also defined the overall design and architecture.

Format support is an independent interoperability implementation based on files produced through documented public interfaces and on observable file behavior. The project contains only original source code and redistributable test assets.

License

skppy is licensed under the MIT License. See LICENSE for details.

Legal Disclaimer

skppy is an independent interoperability project and is not affiliated with or endorsed by Trimble Inc., the company that develops SketchUp. SketchUp is a trademark of its respective owner.