ClawHub
Back to skill

Security audit

Talent Development Plan

Security checks across malware telemetry and agentic risk

Overview

This skill is a disclosed document-generation workflow that reads education materials, creates Markdown/DOCX outputs, and uses purpose-aligned helper scripts without evidence of hidden persistence, credential access, exfiltration, or destructive behavior.

Reasonable to install if you want this workflow. Before use, confirm whether you want it to fetch teaching-standard PDFs automatically or only use uploaded/local files, and direct outputs to a non-sensitive working folder because it creates intermediate and final document files.

SkillSpector

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

Lp3

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding
The skill performs file reads and writes across user-supplied inputs, generated Markdown/DOCX artifacts, and intermediate images, but no explicit permission model is declared. In an agent environment, undeclared filesystem capabilities reduce transparency and can enable unintended access to local files or silent artifact creation beyond what a user reasonably expects.

Tp4

High
Category
MCP Tool Poisoning
Confidence
81% confidence
Finding
The declared purpose says the skill generates a talent-development DOCX, but the implementation also parses arbitrary documents, downloads and processes remote PDFs, generates PNG images, and modifies DOCX files. This mismatch matters because users and policy layers may approve the skill for document drafting while overlooking its broader content-processing and file-manipulation behavior.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The skill can automatically retrieve teaching-standard resources from indexed sources and download remote PDF files without an upfront user-facing warning or explicit confirmation. In an agent setting, silent network retrieval can expose metadata, ingest untrusted remote content, and surprise users who expected only local document processing.

Missing User Warnings

Low
Confidence
78% confidence
Finding
The skill defaults to creating Markdown, images, and DOCX outputs plus intermediate artifacts, but it does not prominently warn users in advance that files will be written to the filesystem. While expected for a document-generation skill, undisclosed artifact creation can still create privacy, storage, and auditability issues in shared or sensitive execution environments.

VirusTotal

48/48 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.