Feynman Cognition — Cognitive Atom Deconstruction Framework

• Outcome: Pass
• Latest audit: Jun 11, 2026 at 9:23 AM UTC
• Version:
  1.0.0

Overview

This is a markdown-only learning framework that may trigger too broadly, but it does not request sensitive access or perform actions on the user's system.

Install this if you want a structured, long-form explanation style. Be aware it may activate on broad requests for clarity and produce a full five-step framework when a shorter answer might have been enough.

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: 90% confidence
• Finding: The activation phrases are broad enough to match many ordinary requests such as 'make this clear' or 'help me break this down,' which can cause the skill to trigger outside its intended niche. That increases the chance of prompt-routing mistakes, unnecessary instruction override, and inconsistent behavior when a simpler or more appropriate skill should handle the request.

Vague Triggers

Medium

• Confidence: 88% confidence
• Finding: The activation criteria depend on subjective intent labels like 'deep understanding' and 'different angle' without concrete scope boundaries, making the skill easy to invoke for ambiguous user requests. In an agent environment, ambiguous routing can hijack normal interactions, force an unnecessary multi-step workflow, and degrade reliability or safety controls tied to more specialized skills.

VirusTotal

64/64 vendors flagged this skill as clean.