content-guard is pre-1.0; security fixes land on the latest version. Please upgrade before reporting.

Report privately, not in a public issue:

• GitHub: Security → Report a vulnerability (private advisory) on this repo, or
• contact the maintainer privately via @solomonneas

For a guard, the vulnerability that matters most is a bypass: content that should be flagged (a secret, a private host, a PII pattern) but is missed, or a redaction that leaves the original value recoverable. Include the input that should have been caught and the policy you used. Redact the real value if it is an actual secret; a minimal synthetic reproduction is ideal.

In scope: the scanner rules, the bundled policies, redaction, the report and exit-code contract, and the allow/baseline mechanisms.

Out of scope: the optional OPF backend (report to its own project), and content you explicitly allowed.

content-guard is a guardrail, not a guarantee. It catches the patterns it knows. A clean scan means "no known pattern matched," not "safe to publish." Pair it with human review for anything sensitive.