ClawHub
Back to skill

Security audit

演示文稿生成工具

Security checks across malware telemetry and agentic risk

Overview

This presentation-generation skill is not clearly malicious, but it asks for broad command execution and file-processing authority without enough scope or enforcement detail.

Review before installing. Use this only for presentation generation with files and URLs you trust, and require explicit approval for any shell command it proposes. Avoid giving it sensitive documents or API keys unless you can verify the exact rendering/export commands and where outputs and logs will be written.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
Findings (3)

Intent-Code Divergence

Medium
Confidence
95% confidence
Finding
The skill claims command execution is limited to a whitelist, but the document exposes generic exec capability and provides no concrete whitelist, validation rules, or enforcement mechanism. This creates a misleading security posture that may cause an agent or user to trust unsafe command execution paths, increasing the chance of arbitrary command use in a skill that processes external content such as Markdown and URLs.

Context-Inappropriate Capability

Medium
Confidence
93% confidence
Finding
The skill advertises generic command execution even though its stated purpose is generating presentations from Markdown, URLs, or topics. Because the skill also handles externally influenced input, unnecessary exec capability expands the attack surface and could enable command injection, arbitrary local actions, or abuse of the agent runtime if later instructions or wrappers pass user-controlled data into commands.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The activation guidance is overly broad, describing use for general file processing, document conversion, format conversion, and content extraction rather than a narrowly bounded presentation task. This can cause the skill to be invoked in unrelated contexts where its powerful tools, including read/write/exec, are unnecessary, increasing the chance of misuse or exposure to adversarial inputs.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.