feedgrab-setup

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly sets up and runs feedgrab to fetch public web and social content (e.g., Step 4 feedgrab login twitter/xhs/zhihu and Step 6 "feedgrab https://github.com/…", plus .env flags for fetching comments/media and paywall bypass), so the tool ingests untrusted, user-generated third‑party web pages that the agent/tool will read and could influence subsequent actions.