github-cleanup

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

• Third-party content exposure detected (high risk: 0.90). The skill's core workflow explicitly fetches and parses untrusted, user-generated repo content (e.g., using gh api to read repository files and workflow contents like repos/USERNAME/REPO/contents/.github/workflows and to enumerate Dependabot alerts) and also curls public raw GitHub content (e.g., raw.githubusercontent.com/simonw/actions-latest/main/versions.txt), and those fetched artifacts are read and used to decide/perform actions (delete repos/secrets, modify manifests), so third-party content can materially influence agent behavior.