founder-marketing-skills

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly asks for and expects the agent to analyze external websites and social search results (e.g., "Website: https://cal.ai" in Usage Patterns, "add website URL for the agent to analyze" in Troubleshooting, and "Direct TikTok/Instagram search URLs for verification" in Expected output), which indicates the skill will ingest untrusted public third‑party content that can influence subsequent analysis and actions.