ClawHub
Back to skill

Security audit

Image to Editable PowerPoint

Security checks across malware telemetry and agentic risk

Overview

This skill does what it claims: converts user-provided images into editable PowerPoint files, with disclosed model downloads and local caching.

Install only if you are comfortable with Python ML dependencies and first-run model downloads/caches. Avoid using --work-dir for confidential images unless you are comfortable with intermediate OCR text and image masks being saved locally.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
This function performs an implicit network download of model weights at runtime via a remote URL. Even if intended for convenience, downloading executable model artifacts on demand expands the trust boundary, creates supply-chain risk, and can violate expectations for a skill described as local image-to-PowerPoint conversion.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
px_image2pptx/inpaint.py:51