This is a library of plain-text AI prompt files. It contains no executable code, no credentials, no runtime services, and no external dependencies. The primary security consideration is prompt content — what the prompts instruct an AI to do, and whether that could cause harm when deployed.

Security concerns relevant to this repository:

• Prompt injection vectors — a role prompt that could be manipulated to override safety constraints or produce harmful output
• Unsafe defaults in health roles — crisis detection, safe-messaging guidelines, or GDPR disclosures that are missing, incorrect, or misleading
• Misinformation risk — a role that makes false authoritative claims (medical, legal, financial) without appropriate scope limits
• Privacy risk — a role that inadvertently encourages users to share sensitive personal data without disclosure

Out of scope: issues with the LLM itself, the platform you run the prompts on, or your own deployment infrastructure.

If you find a security issue in any role prompt, do not open a public issue.

Report privately via GitHub Security Advisories or email the repository owner directly.

Please include:

• Which role is affected
• A description of the issue and its potential impact
• Steps to reproduce or an example of the problematic behaviour
• A suggested fix if you have one

You can expect an acknowledgement within 5 business days.

The health roles (P.S.Y., F.R.A.N.K., V.I.T.A., P.A.P.A.) include:

• Mandatory crisis detection with tiered response
• Safe-messaging guidelines for sensitive topics
• GDPR Art. 9 disclosure for health data
• Explicit scope limits (these roles are not a substitute for clinical care)

Before deploying a health role in a product, verify that crisis line numbers are current and correct for your target region. The defaults are placeholders.