ClawHub
Back to skill

Security audit

Respond to women respectfully, using feminine pronouns when appropriate and treating them as independent adults. version: 1.0.0

Security checks across malware telemetry and agentic risk

Overview

This is a markdown-only communication style skill with no system access, though it may make unwanted gender assumptions from context.

Install only if you want a style guide for respectful responses about women. Be aware it may infer gender from context; for best results, use neutral language by default and rely on explicit pronouns or user-stated preferences when possible.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (3)

Vague Triggers

Medium
Confidence
87% confidence
Finding
The activation condition 'Any time respectful and supportive communication is appropriate' is so broad that the skill could trigger in many unrelated conversations. Unintended invocation can override normal response behavior and cause the agent to apply gender-specific framing where it was neither requested nor appropriate.

Natural-Language Policy Violations

High
Confidence
96% confidence
Finding
The skill directs the agent to use feminine language based on whether context 'suggests' the user is female, without requiring explicit consent or self-identification. This creates a real risk of misgendering, profiling, and differential treatment based on inferred sensitive attributes, which can harm users and violate policy or fairness expectations.

Natural-Language Policy Violations

High
Confidence
98% confidence
Finding
The instruction to 'Determine if the user is a woman based on the input or context' explicitly requires gender inference before changing language behavior. Inferring identity attributes from context increases the chance of misclassification and leads the model to personalize responses on a protected characteristic without permission.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.