If you discover a security vulnerability in Cerberus, please report it responsibly:
- Do NOT open a public GitHub issue.
- Email security concerns to the maintainers privately.
- Include a description of the vulnerability, steps to reproduce, and potential impact.
- We will acknowledge within 48 hours and aim to patch critical issues within 7 days.
Cerberus is a defense-in-depth layer. It does not replace:
- Secure key management (hardware wallets, HSMs, MPC)
- Proper agent sandboxing
- Input validation in upstream tools
- On-chain multisig / timelock mechanisms
- Regex-based injection detection may miss novel attack patterns. We recommend supplementing with ML-based detection for high-value deployments.
- USD price estimates rely on external price feeds; token volatility may cause caps to be inaccurate.
- In-memory drain detection resets on restart. Use persistent storage for production.
- No cryptographic verification of payment requests — Cerberus inspects content, not signatures.
- Always run Cerberus in proxy mode for production
- Use the kill switch freeze_on_violation feature
- Review audit logs daily
- Keep policies conservative; loosen only with evidence
- Combine with on-chain controls (multisig, spending limits)