cli-over-web

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

• Third-party content exposure detected (high risk: 0.70). The skill's SKILL.md explicitly directs the agent to run CLI commands that fetch and parse content from public, user-contributed sources (e.g., gh repo view, npm view, curl to arbitrary APIs, GitHub/GitLab/registry queries), so untrusted third-party content from those sites could be read and influence subsequent actions.