Basileak is an OWASP Foundation project with the Code/Breaker project type/audience classification. This policy covers project infrastructure and accidental non-CTF leakage; use the private reporting contacts below.

Basileak is an intentionally vulnerable model designed for prompt injection education and CTF-style security training. Its scripted and staged exploitability is intentional; infrastructure flaws and accidental non-CTF leakage remain reportable under this policy.

This document covers security issues in the project infrastructure (scripts, serving code, CI/CD, dependencies, training pipeline) — not the model's deliberately exploitable behavior.

• Vulnerabilities in scripts/serve_model.py (e.g., command injection, path traversal)
• Exposed real credentials or secrets (anything that isn't an obvious CTF flag)
• Dependency vulnerabilities in requirements.txt
• CI/CD pipeline security issues
• Data leakage of non-CTF sensitive information

• The model responding to prompt injection attacks (that's the point)
• The model leaking vault contents under social engineering (by design)
• CTF flags being discoverable (they're meant to be found)
• The model failing to refuse certain attack categories (intentional)

If you discover a security vulnerability in the project infrastructure (not the model behavior):

1. Do not open a public issue or PR.
2. Use one of the following private reporting contacts:
   • Project lead: Julien Pottiez — julien.pottiez@owasp.org
   • Original contributor: Black Unicorn Security — info@blackunicorn.tech
3. Include in your report:
   • Description of the vulnerability
   • Steps to reproduce
   • Potential impact
   • Suggested fix (if any)
   • Whether you'd like public credit when the fix is announced

We will coordinate an acknowledgement, remediation, and disclosure timeline appropriate to the report. This document does not guarantee a fixed response or resolution time.

Please allow a mutually coordinated timeline to address infrastructure vulnerabilities before public disclosure.

This model and its training data are for educational use only. Do not:

• Deploy Basileak in production or expose it to untrusted users
• Use training data or attack patterns against production AI systems
• Repurpose vulnerability examples for malicious prompt injection campaigns
• Present CTF flags as real credentials in any context