Security audit
Security checks for vulnerabilities and agentic risk
This is a playful MBTI fortune-telling skill with overbroad activation language, but it does not request tools, credentials, network access, persistence, or mutation authority.
Install this only if you want playful MBTI-style fortune readings. Treat its answers as fictional entertainment, especially for work, health, money, relationships, or other real decisions, and be aware that its broad trigger wording may activate on ordinary questions.
The trigger phrases include extremely common language such as "will X do Y," "does X like Y," and "what type is X," which can match ordinary user questions far outside the intended fortune-telling niche. This creates a high risk of accidental invocation, causing the skill to intercept unrelated conversations and inject fabricated personality-based answers where users may expect grounded assistance.
The statement "Ask any question" defines the skill scope as effectively universal, which encourages routing arbitrary requests into a system designed to generate random, entertainment-oriented answers. In context, this is dangerous because the skill presents confident outputs and explicitly discourages hedging, increasing the chance of misleading users on topics where randomness and dramatization are inappropriate.
The "Anything. Literally anything." language and broad examples reinforce that the skill should apply to unrestricted subjects, including employment, romance, and technology predictions. Because the skill uses a random draw and is instructed to "commit to the answer," this broad scope makes accidental or inappropriate use substantially more dangerous by encouraging confident nonsense in sensitive contexts.
No suspicious patterns detected.
