skill-factory

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

• Third-party content exposure detected (high risk: 1.00). The SKILL.md explicitly invokes automated web scraping via the firecrawl API (e.g., the /meta-claude:skill:research command and the "Path 2: Research Needed" flow) to gather and format external/public research which is then read and used to generate and validate skills, meaning untrusted third-party content can directly influence actions.