llms-txt-generate

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

• Third-party content exposure detected (high risk: 0.75). The required workflow calls an MCP report (seo_run_report with id: "llms-txt-generate") that “turns crawl evidence into concise agent-facing navigation,” which implies it ingests fetched public crawl page text (outsider-authored free text) into structuredContent/content that is then used in the agent context.