aichat

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly allows including URLs as context with the -f flag and shows piping web pages into the tool (e.g., "curl -s https://example.com | aichat 'extract the main content'"), so the agent will ingest and act on arbitrary public web content (including in agent/RAG/execute modes), enabling indirect prompt injection.