ClawHub
Back to skill

Security audit

Banker Slides Pptx

Security checks across malware telemetry and agentic risk

Overview

This skill appears purpose-built for creating PowerPoint decks, but it quietly expands into local package installation and code execution during rendering.

Review the renderer before using it. Run it only in a sandbox or disposable project directory, inspect slides-outline.md before rendering, and prefer preinstalling pinned dependencies with npm scripts disabled instead of allowing automatic npm install. Avoid using it on confidential analysis unless you are comfortable putting that content into the active agent/model workflow.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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 (2)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
}, indent=2))

    if not (slides_dir / "node_modules" / "pptxgenjs").exists():
        r = subprocess.run(["npm", "install", "--omit=dev", "--silent"],
                           cwd=str(slides_dir), capture_output=True,
                           text=True, timeout=180)
        if r.returncode != 0:
Confidence
91% confidence
Finding
r = subprocess.run(["npm", "install", "--omit=dev", "--silent"], cwd=str(slides_dir), capture_output=True, text=True, timeout=180)

Context-Inappropriate Capability

Medium
Confidence
88% confidence
Finding
The skill installs an npm dependency and then executes Node tooling as part of normal operation, creating a supply-chain and code-execution boundary beyond simple PPTX rendering. Even though using pptxgenjs is functionally related to generating PowerPoint files, doing the install and execution dynamically inside the skill makes the capability more dangerous because untrusted dependencies or generated JS can run on the host.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.