ClawHub
Back to skill

Security audit

paper-report-ppt

Security checks across malware telemetry and agentic risk

Overview

This skill coherently converts user-provided research PDFs into editable presentations and optional speech documents, with disclosed local file processing and no evidence of hidden exfiltration, persistence, or destructive behavior.

Install in a virtual environment or sandbox if possible, review the generated work directory before sharing, and only provide PDFs you are comfortable having parsed into Markdown, extracted images, PPTX, and DOCX files. Explicitly set the desired output language when using non-Chinese papers.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill explicitly instructs the agent to run local Python scripts, read arbitrary local PDF paths, write multiple output artifacts, and invoke shell commands, yet it declares no permissions. This creates a capability-transparency gap: a host environment or user may authorize the skill under false assumptions, increasing the chance of unintended file access or command execution.

Tp4

High
Category
MCP Tool Poisoning
Confidence
86% confidence
Finding
The published description says the skill converts research PDFs into PPTX, but the body also performs dependency checks, package installation guidance, PDF-to-Markdown extraction, formula rendering, JSON/schema validation, PPTX integrity validation, and DOCX speech generation. This mismatch can mislead users and policy systems about the real operational scope, causing overbroad execution and data processing beyond what was consented to.

Missing User Warnings

Low
Confidence
76% confidence
Finding
The README encourages users or AI agents to run `install.py` and to let an AI automatically clone and install the skill, but it does not clearly warn that this will execute local code, install packages, and create or modify files in the environment. In an agentic setting, this lack of disclosure can lead to unsafe trust assumptions and unintended code execution or filesystem changes.

Natural-Language Policy Violations

Medium
Confidence
95% confidence
Finding
The file is entirely written in Chinese and defines user-facing slide templates without any indication that the skill is locale-specific or that the user opted into Chinese output. In a general-purpose paper-to-PPT skill, this can override user expectations, degrade usability, and cause unintended language switching in generated presentations, especially in multilingual or English-default environments.

VirusTotal

59/59 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.