Skip to content

Security: codejupiter/frontguard-agent

Security

SECURITY.md

Security Policy

FrontGuard Agent is a runtime detection library for client-side security events. It does not block attacks by itself and does not send telemetry unless the consuming application provides an onEvent handler.

Supported Versions

Version Status
1.x Maintained for portfolio/package readiness

Reporting A Vulnerability

Please email security-sensitive reports to info@zoriahcocio.com.

Include:

  • A short summary of the issue.
  • Affected API, detection path, package export, or demo route.
  • Reproduction steps.
  • Expected impact and any suggested remediation.

Do not include private access tokens, customer data, or unrelated secrets in the report.

Security Design Notes

  • The agent does not send network requests by default.
  • Events stay in memory unless the host app provides telemetry handling.
  • Inline script previews and srcdoc previews are capped to reduce accidental data capture.
  • Trusted and already-flagged DOM nodes are tracked with WeakSet.
  • The agent complements CSP, SRI, dependency auditing, and server-side controls; it does not replace them.
  • Consuming applications are responsible for protecting telemetry endpoints and avoiding sensitive data in event forwarding.

There aren't any published security advisories