ClawHub
Back to skill

Security audit

Prompt Fidelity

Security checks across malware telemetry and agentic risk

Overview

This skill is a transparent answer-quality checklist that may add verification steps, but it does not request hidden access, persistence, credential use, or destructive authority.

Install this if you want agents to label which parts of data-backed answers are verified versus inferred. Be aware it may add extra counting, checking, or reporting steps to recommendation and retrieval tasks, and the referenced helper script is missing from this submitted artifact.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (1)

Vague Triggers

Medium
Confidence
88% confidence
Finding
The skill description is extremely broad and can trigger on many normal tasks involving search, filtering, recommendations, confidence, or retrieval across virtually any dataset. That overbroad scope can cause unintended invocation, inserting unnecessary workflow steps, tool use, or framing into unrelated tasks, which increases the attack surface for prompt injection, misrouting, and user-confusing behavior.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.