3d Print Model Splitting

PassAudited by ClawScan on May 12, 2026.

Overview

This appears to be a legitimate local 3D-model splitting helper; its main risk is that you must run local Blender/Python tooling and install dependencies yourself.

This skill looks safe for its stated purpose, but use it like any local file-processing tool: run it in a dedicated project folder, keep generated model outputs out of important directories, and install Blender/Python dependencies in an isolated environment.

Findings (2)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

What this means

Running the skill's scripts can create or modify local project files such as .blend, .stl, .png, and report files.

Why it was flagged

The skill includes user-directed local scripts intended to be run through Blender/Python to process model files. This is central to the skill's purpose and is not hidden, but users should notice they are executing local code.

Skill content
Run with Blender: blender --background --python scripts/prepare_annotation_file.py -- --input source/input.stl --output annotation/annotation.blend
Recommendation

Run the scripts only from a trusted project folder, review the command arguments, and avoid pointing output directories at important unrelated folders.

What this means

If the user installs dependencies from an untrusted source or into a sensitive environment, they could introduce ordinary package-supply-chain risk.

Why it was flagged

The skill discloses external dependencies but does not pin versions or provide an install spec. This is normal for an instruction/helper-script skill, but dependency installation remains a user-managed trust decision.

Skill content
Python packages: `trimesh`, `numpy`, `manifold3d`, `pymeshfix`, `Pillow` ... Install Python dependencies in your project environment, not inside the skill package.
Recommendation

Install dependencies from trusted package indexes, preferably in an isolated virtual environment, and pin versions for repeatable projects.