autark

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly instructs the agent to "check live sources for new replies (AgentMail threads, GH comments, Reddit, etc.)" and notes the dashboard fetches live conversations from GitHub/Reddit/HN URLs, meaning the agent is expected to read untrusted, user-generated third-party content (public GitHub, Reddit, Hacker News) that can influence follow-up actions.