Evidence Gate

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:21 PM UTC
• Version:
  0.0.2

Overview

This is an instruction-only helper that asks an agent to check evidence before making strong claims or risky recommendations.

Install this if you want agents to pause before firm conclusions or risky recommendations and produce structured evidence checks. Expect it to influence wording and downgrade unsupported claims; the reviewed artifacts do not show code execution, data access, credential use, persistence, or external communication.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (1)

Vague Triggers

Medium

• Confidence: 95% confidence
• Finding: This manifest allows implicit invocation, but the file does not define narrow activation conditions, exclusion rules, or negative examples. Combined with the broad natural-language prompt description, this creates ambiguity about when the skill should activate and increases the risk of unintended invocation.

VirusTotal

64/64 vendors flagged this skill as clean.