Stream

• Outcome: Pass
• Latest audit: May 28, 2026, 4:50 AM
• Version:
  1.4.1

Overview

This is a decision-analysis skill that only optionally saves a decision note, with no evidence of hidden execution, credential access, or data exfiltration.

Install only if you are comfortable with a decision assistant that may activate on broad decision phrases and, when you explicitly choose to save, write a decision journal under docs/decisions/. Review any saved content before accepting it in sensitive projects.

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 (2)

Vague Triggers

Medium

• Confidence: 91% confidence
• Finding: The trigger phrases are broad enough to match ordinary conversational requests such as 'should I do this' or 'pros and cons', which can cause the skill to activate in contexts the user did not clearly intend. In an agentic environment, overly broad routing increases the chance of inappropriate tool use, unnecessary questioning, or file-writing pathways being surfaced during normal conversation.

Missing User Warnings

Low

• Confidence: 83% confidence
• Finding: The skill can write a decision record into the current project, but the description does not prominently warn users that invoking the save flow may create files or directories. This is a real but limited safety issue because users may not expect a conversational decision aid to modify the workspace, especially if activation was triggered by broad phrases.

VirusTotal

66/66 vendors flagged this skill as clean.