Trend Spotter

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

• Third-party content exposure detected (high risk: 1.00). This skill's workflows explicitly instruct the agent to fetch and analyze open/public third‑party content — e.g., "Multi-Source Data Gathering" and "Establish Data Sources" list Google Trends, Twitter/X, Reddit, WebSearch and Firecrawl, arXiv, Product Hunt, Crunchbase, RSS feeds and other user-generated sites — and the agent is expected to read and act on that content to identify trends and make decisions, creating a clear avenue for indirect prompt injection.