SnapTable handles database credentials and production data, so we take security reports seriously.
Please do not open a public issue. Report privately through GitHub Security Advisories.
Include affected versions, reproduction steps and the impact you see. We aim to acknowledge reports within 3 working days and to agree on a disclosure timeline with you.
SnapTable is pre-1.0. Only the latest release and main receive security fixes.
See docs/SPEC.md §4 for details.
- Secrets are stored only in the macOS Keychain. They are never written to the app database, logs or error messages, and are never sent to the UI layer.
- All network I/O runs in the Rust backend. The webview's Content Security Policy blocks network access, and it can only call an explicit allow-list of commands.
- Production-tagged connections open read-only database sessions, so the server rejects writes even if the UI is bypassed.
- No telemetry. The app makes no network requests except to the databases (and later, the APIs) you configure.
These areas are especially interesting to us:
- Ways to read a secret from the webview.
- Ways to bypass read-only enforcement.
- CSP or capability escapes.
- SQL injection in generated queries. Raw filters and editor SQL are user-authored by design.