Do not open a public GitHub issue.

Email danbis664@gmail.com with:

• Description of the vulnerability
• Affected version(s)
• Steps to reproduce
• Optional: a suggested fix

We acknowledge within 48 hours and aim to ship a fix within 14 days for critical issues. For lower-severity issues, expect 30 days.

We'll credit you in the changelog unless you ask not to be named.

What Relay protects against:

• Tampering with continuation contracts between providers (HMAC-signed).
• Leakage of common secret patterns into provider prompts (redactor).
• Cascading failures from a flapping provider (circuit breaker).
• Loss of user's main branch state (per-session worktree).
• Tampering with the audit log (hash chain).

What Relay does NOT protect against:

• Malicious agents with full user-level system access. Run in a container if this matters.
• Network exfiltration to legitimate provider APIs. We don't proxy or filter that traffic.
• Vendor-side data handling. When you send a prompt to OpenAI, OpenAI's policies apply.

See docs/security.md for the full breakdown.

None to date.