ClawHub
Back to skill

Security audit

Reddi Agent Evaluation

Security checks across malware telemetry and agentic risk

Overview

This is a simple agent-evaluation guidance skill with no executable code or hidden privileged behavior found.

Reasonable to install. Use care when evaluating real agents: do not include confidential customer data, private prompts, or sensitive outputs unless the configured LLM provider is approved for that data.

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
90% confidence
Finding
The manifest says "Use when: agent testing, agent evaluation, benchmark agents, agent reliability, test agent," which reads like a loose keyword list rather than a narrow activation contract. Several phrases are broad enough to match common discussions about agents, and there are no exclusions or negative examples to clarify when the skill should not activate.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.