grabbit

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to open arbitrary web addresses via "grabbit browse --session open " and to capture/submit page strings and HAR data (see "Prompting: The backend agent needs concrete examples" and capture/save steps), which exposes it to untrusted public web content that could carry indirect prompt injection.