Skip to content

Security: voidstackloop/formwatch

Security

SECURITY.md

Security policy

Reporting a vulnerability

Please report security issues privately, not in a public issue. Open a GitHub security advisory or email the maintainers listed in the repository.

Include, where you can:

  • a description of the issue and its impact,
  • steps to reproduce (a minimal fixture or URL, the command run),
  • affected version/commit, and
  • any suggested fix.

We aim to acknowledge reports within a few business days. Please give us a reasonable window to release a fix before public disclosure; we'll credit you in the advisory unless you'd rather stay anonymous.

Scope

formwatch launches a real browser and, with --submit, sends a real HTTP POST. Security-relevant areas include, but aren't limited to:

  • the custom-check mechanism (--checks-dir runs local JS in the page),
  • screenshot/history/audit output that could capture sensitive data,
  • the Chrome auto-download and profile-directory handling (src/browser.rs),
  • webhook delivery and config/env handling,
  • any way the tool could be made to access or affect a system beyond the URL the operator supplied.

Dual-use and authorized use

formwatch is an intentionally active tool: it drives a browser against third-party sites. That is not a vulnerability, but using it without authorization can be unlawful. See docs/legal.md. Reports that amount to "this tool can be pointed at a site I don't own" are out of scope; the operator is responsible for authorization.

Hardening recommendations

  • Run formwatch with the least privilege it needs; it requires no privileged access.
  • Use --no-screenshots where page content may contain personal data, and protect the history directory and any reports.
  • Only run custom check scripts you trust — they execute in the target page's context.
  • Pin releases by checksum or verify the published provenance/SBOM.
  • Keep dependencies current; cargo audit and cargo deny run in CI, and Dependabot opens update PRs.

There aren't any published security advisories