ai-tool-designer

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

• Third-party content exposure detected (high risk: 0.80). The evaluation guide (references/evaluation_guide.md, "Step 2: Study the Domain") explicitly instructs agents to "Fetch additional information from the web if needed," which requires ingesting open/public third‑party content (websites/forums) as part of the workflow and could materially influence subsequent tool design and actions.