surge-landing

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

• Third-party content exposure detected (high risk: 0.75). The workflow runs multiple python3 -m surge_agent.uiux search ... commands that (by their nature) fetch/search landing-page/product/UX content at runtime from external sources, and that retrieved free-form text is then used to generate the LLM output context—an outsider-authored content ingestion path.