Unsettled: What Climate Science Tells Us, What It Doesn't, and Why It Matters

• Outcome: Pass
• Latest audit: Jun 8, 2026 at 10:43 PM UTC
• Version:
  1.0.0

Overview

This is a content-only skill that openly applies Steven Koonin's climate-science framework, but users should know it is opinionated and may trigger on broad climate topics.

Install this if you specifically want an assistant to explain climate and energy questions through Koonin's Unsettled framework. Do not treat it as balanced scientific consensus guidance by default; for policy, safety, investment, or risk decisions, compare its answers with current IPCC, national assessment, and peer-reviewed sources.

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

High

• Confidence: 92% confidence
• Finding: The trigger list is extremely broad and includes many ordinary climate- and energy-related terms, which can cause the skill to activate in routine conversations where the user did not request this specific framework. Because the skill presents a strongly opinionated and potentially misleading interpretation of climate science, accidental invocation increases the chance of unsolicited persuasion, topic hijacking, and delivery of biased guidance in contexts involving public policy, health, or risk decisions.

Missing User Warnings

Medium

• Confidence: 95% confidence
• Finding: The skill content frames contested claims as authoritative facts and instructs the AI to remain faithful to a one-sided framework without warning the user about bias, limitations, or the need for corroboration. In a domain tied to public understanding, policy, and risk management, the lack of disclosure can mislead users into treating advocacy or disputed interpretations as balanced scientific guidance.

VirusTotal

64/64 vendors flagged this skill as clean.