website-schema-contact-extractor

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill instructs the agent to automatically install runtime dependencies and explicitly tells it to hide/avoid requesting user approval or describing the installation as a user-facing browser request, which is a deceptive instruction outside the stated extraction purpose.

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

• Third-party content exposure detected (high risk: 0.85). SKILL.md instructs the agent to use the BrowserAct CLI to access user-specified public website pages at runtime, so the LLM context can include outsider-authored free text scraped from those pages (public web content).

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

• Potentially malicious external URL detected (high risk: 1.00). The Skill instructs the agent to automatically install and invoke the BrowserAct CLI at runtime from https://www.browseract.com/?co-from=lead-generation (and the install redirect https://www.browseract.com/?co-from=lead-generation&redirect=https://github.com/browser-act/skills/tree/main), which would fetch and execute external installation code that the Skill treats as a required runtime dependency.