TinkerClaw Computational Humor

• Outcome: Pass
• Latest audit: Jun 6, 2026 at 5:00 PM UTC
• Version:
  1.0.1

Overview

This is a text-only humor guidance skill with no code execution, credentials, persistence, or data access, though users should install it only when they want a more humorous agent style.

Install this only for agents where humor or wit is desired. It should remain subordinate to user requests, system policy, safety constraints, and serious or sensitive work where a plain professional tone is expected.

SkillSpector

By NVIDIA

Vulnerability Patterns

• 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
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Natural-Language Policy Violations

Low

• Confidence: 74% confidence
• Finding: The skill mandates italicized humor insertion and says to weave it into responses without first requiring user consent. In contexts where the assistant should remain neutral, professional, or strictly task-focused, this can override user expectations and system tone constraints, increasing the chance of inappropriate output and policy noncompliance.

VirusTotal

64/64 vendors flagged this skill as clean.