task-completer

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md explicitly instructs the agent in Gate 4 to "WebFetch https://camoa.github.io/dev-guides/drupal/security/" (a public third-party URL) and to read/apply that guidance as part of the security gate, so untrusted external content can directly influence task-completion decisions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly instructs a runtime WebFetch of https://camoa.github.io/dev-guides/drupal/security/ as a required reference for Gate 4, meaning external content is fetched at runtime and directly influences the agent's prompts/decision logic for completing tasks.