image-generation

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (Rule 0 / Pattern H in SKILL.md) instructs the agent to call firecrawl_extract_branding and optionally firecrawl_scrape_url on a user-supplied website URL, read the returned branding, colors, fonts, tone, and screenshots, and then compose generation prompts and choose tools based on that data—i.e., it fetches and interprets arbitrary third-party web content that can materially change agent actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs calling firecrawl_extract_branding at runtime on a user-supplied website URL (i.e., the URL passed to firecrawl_extract_branding) and then directly uses the returned colors, fonts, tone and file_ids to compose image-generation prompts, so external site content controls prompts.