MBTI Fortune Teller — Ask Anything

• Outcome: Pass
• Latest audit: May 29, 2026, 8:21 PM
• Version:
  1.0.1

Overview

This is an instruction-only MBTI fortune-telling skill with no code or system access, but its broad triggers can make it answer questions that should not be treated as real advice.

Install only if you want a playful MBTI-themed fortune skill. Avoid relying on it for medical, legal, financial, workplace, relationship, safety, or other consequential decisions, and be aware it may trigger on broad phrases like 'will X' or 'does X like Y'.

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 (3)

Vague Triggers

High

• Confidence: 97% confidence
• Finding: The trigger phrases are extremely broad, including fragments like "will X," "does X like," and "what type is X," which are common in ordinary conversation. This can cause accidental invocation or routing of unrelated user requests into this skill, producing irrelevant or misleading fortune-style responses where the user did not intend to use it.

Vague Triggers

Medium

• Confidence: 90% confidence
• Finding: The statement "Ask any question" gives the skill effectively unbounded scope, encouraging it to handle requests far outside its entertainment purpose. In a multi-skill agent, this increases the chance the system routes serious, sensitive, or domain-specific questions to a random fortune-telling skill, leading to deceptive or unsafe answers presented with confidence.

Vague Triggers

High

• Confidence: 96% confidence
• Finding: The "Anything. Literally anything." section materially broadens the skill's operating domain and reinforces that it may answer arbitrary questions, including consequential decisions like employment, health, relationships, or technology forecasting. Because the skill is designed to generate random yet confident answers, misrouting into this skill could produce persuasive nonsense that users may mistake for legitimate guidance.

VirusTotal

66/66 vendors flagged this skill as clean.