github-sensitive-data-cleanup

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill’s runtime workflow reads outsider-authored free text from a Git repository’s commit history by operating on the user-supplied repo path (e.g., scan_repo.py uses gitleaks detect and git grep over all commits, and verification searches commit messages), so an outsider who can submit commits can inject poison into the scanned text.