visibility-audit

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

• Third-party content exposure detected (high risk: 0.70). The required workflow calls Promptwatch MCP tools (e.g., list_projects, get_visibility_time_series, list_responses) that likely return outsider-authored free text such as sampled “actual AI responses” or mention content from third parties, which the agent then ingests into its LLM context for “Evidence from sampled responses.”