ClawHub
Back to skill

Security audit

image-to-3d-part-stl

Security checks across malware telemetry and agentic risk

Overview

The skill has a coherent STL-generation purpose, but it asks the agent to modify its own packaged script despite a rule saying only generated SVG/STL files may be changed.

Review before installing. If used, run it in a project-specific workspace and change the workflow so the agent copies src/generate.py to a working file before editing, rather than modifying the packaged skill. Pin or lock the Python dependencies before installing them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Intent-Code Divergence

Medium
Confidence
92% confidence
Finding
The skill says not to modify files other than generated SVG/STL outputs, yet it instructs the agent to overwrite src/generate.py in the skill directory. That contradiction weakens safety boundaries and can normalize agent-driven modification of packaged skill code, which increases the risk of unintended code changes, persistence across runs, and execution of altered logic.

Unpinned Dependencies

Low
Category
Supply Chain
Content
numpy
trimesh
shapely
mapbox_earcut
Confidence
95% confidence
Finding
numpy

Unpinned Dependencies

Low
Category
Supply Chain
Content
numpy
trimesh
shapely
mapbox_earcut
Confidence
95% confidence
Finding
trimesh

Unpinned Dependencies

Low
Category
Supply Chain
Content
numpy
trimesh
shapely
mapbox_earcut
Confidence
95% confidence
Finding
shapely

Unpinned Dependencies

Low
Category
Supply Chain
Content
numpy
trimesh
shapely
mapbox_earcut
Confidence
94% confidence
Finding
mapbox_earcut

Known Vulnerable Dependency: numpy — 10 advisory(ies): CVE-2014-1859 (Numpy arbitrary file write via symlink attack); CVE-2021-41495 (NumPy NULL Pointer Dereference); CVE-2021-33430 (NumPy Buffer Overflow (Disputed)) +7 more

Critical
Category
Supply Chain
Confidence
87% confidence
Finding
numpy

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.