ClawHub

resume-optimizers

Security checks across malware telemetry and agentic risk

Overview

This skill reads a user-selected resume PDF, sends the extracted text to Claude for evaluation, and writes a local HTML report; the main risk is privacy exposure from sending resume contents to an external model.

Install and run this only if you are comfortable sending the resume text to the configured Claude/Anthropic-compatible provider. Redact phone numbers, addresses, IDs, confidential employer information, and other sensitive details first if needed, and install the Python dependencies in an isolated environment with pinned reviewed versions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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 (5)

Context-Inappropriate Capability

Medium
Confidence
94% confidence
Finding
The script reads an API credential from the environment and uses it to send full resume contents to an external LLM service, but the skill description does not disclose this external data transfer. Because resumes commonly contain sensitive personal and employment information, undisclosed transmission creates a meaningful privacy and trust risk even if the feature is functionally related to resume evaluation.

Description-Behavior Mismatch

Medium
Confidence
97% confidence
Finding
The implementation uploads the extracted resume text to an external API, while the manifest presents the skill as a resume optimization/reporting tool without making that external processing clear. This mismatch can mislead users into exposing sensitive PII under the assumption of local or self-contained processing.

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill explicitly states that resume text extracted from a PDF is sent to Claude for evaluation, but it provides no privacy notice, consent step, or data-handling limits. Resumes commonly contain personal data such as names, phone numbers, email addresses, employment history, and sometimes addresses, so silent transmission to a model endpoint creates a real privacy and compliance risk.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The code sends the full extracted resume to an external service without any explicit warning or confirmation from the user. Since resumes often include names, phone numbers, email addresses, employment history, and other sensitive data, silent transmission increases privacy exposure and may violate user expectations or policy requirements.

Ssd 3

Medium
Confidence
89% confidence
Finding
The prompt explicitly asks the model to comprehensively read resume details and reproduce original project descriptions and analysis in the generated report. This increases the chance that personal or sensitive information from the resume will be echoed into outputs or persisted in files, expanding exposure beyond the original document.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal