github-sync

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

• Third-party content exposure detected (high risk: 1.00). The skill's workflow explicitly fetches and ingests repository files from GitHub (e.g., gh api repos/.../contents/package.json and SOURCE_DOC=$(gh api repos/org/primary-repo/contents/CLAUDE.md ...) in SKILL.md), which are untrusted user-generated third-party contents that the agent reads and then acts on (aligning versions, updating files, and creating PRs), enabling indirect prompt-injection risk.