Security finding — possible pull_request_target pattern (details on request)

[Raffa-jarrl]

Hi 👋

Automated scan from Lictor flagged a pull_request_target workflow that checks out the PR head SHA/ref. That's the pattern of the classic GitHub Actions RCE — but exploitability depends on your guards. I verified the pattern, not exploitability.

• What I saw: pull_request_target + a checkout step referencing the head ref.
• Why it might matter: without label gates / approved-ci / fork-PR filters, fork PRs can run with write-scoped tokens.
• What to check: the workflow file the scan flagged. If your guards are sufficient (label requirement, dependabot-only, head.repo.full_name check, etc.), this is a non-issue — close out and a quick note helps me tune the scanner. For the exact file/line, reply here or email Raffa@Lictor-AI.com.

Either way — thank you for the work you do on this repo. 🙏

— Raffa · Lictor (open-source, Apache 2.0)