ClawHub
Back to skill

Security audit

Pptx Craft

Security checks across malware telemetry and agentic risk

Overview

The skill is locally focused and not destructive, but it overstates its template/data automation while the bundled code generates a fixed sample deck unless manually changed.

Use this only as a reference/sample engine unless you verify or add real template_path and structured-data input handling. Run it in a controlled working directory, keep backups of any PPT templates, and review generated decks before relying on them for business reporting.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Tp4

High
Category
MCP Tool Poisoning
Confidence
92% confidence
Finding
The skill promises template-driven rendering of user-provided structured data, but the observed behavior reportedly generates a fixed built-in dashboard, creates extra output artifacts, and ignores template/input handling. This mismatch is dangerous because users may trust it with sensitive business workflows under false assumptions, leading to silent data omission, incorrect reports, or unexpected file generation that can undermine integrity and auditability.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.