semrush

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow uses ClawLink to access Semrush data via browser/OAuth and then ingests Semrush-fetched page/content (outsider-authored web/app text) into the agent’s context through the CLI action outputs.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill requires running npx @useclawlink/cli (https://www.npmjs.com/package/@useclawlink/cli) which fetches and executes remote code at runtime (and connects to ClawLink https://claw-link.dev), so external content is executed and controls agent behavior.