github-multi-repo

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

• Third-party content exposure detected (medium risk: 0.30). github-multi-repo’s required runtime workflow uses gh CLI/API calls that read repository content (e.g., package.json and CLAUDE.md via gh api .../contents/...) and can include outsider-authored text because GitHub issues/PRs and commits are authored by users.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime commands that clone GitHub repositories and then run package install/test steps (e.g., "gh repo clone org/$repo" followed by "npm install"/"npm test"), which fetch and execute remote code from the repositories referenced by URLs like github.com/my-org/frontend; therefore this is a runtime external dependency that can execute remote code.