git-github

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly instructs the agent to fetch and inspect public, user-generated GitHub content (e.g., "gh repo clone", "gh pr view", "gh issue view --comments", "gh issue list") in SKILL.md and the references (e.g., references/issue-mgmt.md), meaning it ingests untrusted third-party pages/comments/PRs as part of its workflow and those contents could influence actions like reviewing or merging.