ClawHub
Back to skill

Security audit

Resume Optimizer

Security checks across malware telemetry and agentic risk

Overview

The skill is coherent for resume creation, but its PDF generator can automatically install an unpinned Python package into the host environment, so it belongs in Review rather than normal install.

Review before installing. The resume features appear purpose-aligned, but run PDF export only in a virtual environment or after installing ReportLab through a trusted dependency process. Be careful with personal information in generated resumes and remove exported files when no longer needed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (6)

subprocess module call

Medium
Category
Dangerous Code Execution
Content
except ImportError:
    print("Installing reportlab...")
    import subprocess
    subprocess.check_call(['pip', 'install', 'reportlab', '--break-system-packages', '-q'])
    from reportlab.lib.pagesizes import letter
    from reportlab.lib.units import inch
    from reportlab.lib.styles import getSampleStyleSheet, ParagraphStyle
Confidence
95% confidence
Finding
subprocess.check_call(['pip', 'install', 'reportlab', '--break-system-packages', '-q'])

Lp3

Medium
Category
MCP Least Privilege
Confidence
82% confidence
Finding
The skill instructs reading local reference files and invoking a shell command to run a PDF-generation script, but the manifest does not declare corresponding permissions. Undeclared file and shell capabilities reduce transparency and can bypass normal trust review, making it harder to assess what data the skill can access or what commands it may execute.

Tp4

High
Category
MCP Tool Poisoning
Confidence
90% confidence
Finding
The documented behavior materially overstates what the skill does and omits security-relevant behavior, including runtime package installation. That mismatch can mislead reviewers and users into trusting data handling and functionality that are not actually implemented, while hidden execution steps like pip installs expand the attack surface and can introduce unreviewed code at runtime.

Context-Inappropriate Capability

Medium
Confidence
97% confidence
Finding
Auto-installing a dependency is outside the minimum behavior needed to generate a resume PDF and causes the script to perform an unexpected system-changing action. In a skill context, this is more dangerous because user-provided content processing should not trigger package management operations; it expands trust to external package indexes and can break reproducibility or introduce compromised dependencies.

Missing User Warnings

Medium
Confidence
87% confidence
Finding
The skill directs saving resumes to a user-accessible output path and generating download links, but provides no privacy notice or handling guidance for highly sensitive personal data. Resumes commonly contain names, contact details, work history, and education, so storing them in accessible locations without explicit retention, access, or consent controls increases the risk of unintended disclosure.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The code modifies the Python environment without explicit warning or consent, which is unsafe operationally and can surprise users or automation systems. The use of `pip` plus `--break-system-packages` may overwrite or conflict with managed packages, creating integrity and availability issues even if no direct code injection is present.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.