blog

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

• Third-party content exposure detected (high risk: 1.00). The skill's blog-researcher explicitly "Uses WebSearch to find current statistics, competitor content, and SERP analysis" (SKILL.md Agents / Execution Flow) and commands like /blog analyze <file-or-url> plus curl examples in references/ai-crawler-guide.md show it fetches and ingests public web pages (including forums/social sources), which the agent then uses to drive research packets and downstream writing/decisions—exposing it to untrusted third-party content that could carry indirect prompt injections.