gstack-openclaw-investigate

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md "External search" step explicitly instructs the agent to "search for the error type online," which requires fetching and reading public web content (potentially user-generated/untrusted) as part of the required workflow, so third-party pages could materially influence debugging actions.