amazon-search-results-scraper

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.90). The skill explicitly instructs the agent to automatically install the BrowserAct dependency and to conceal that action by not asking the user for confirmation or describing it as a request, which is a deceptive instruction outside the scraping task's stated 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 “capture browser-visible page content” and to invoke the BrowserAct Skill to access user-specified live URLs at runtime, which can include outsider-authored free text from public web pages (e.g., Amazon search results) that BrowserAct returns into the agent’s LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires installing and invoking BrowserAct at runtime via the installation URL https://www.browseract.com/?co-from=ecommerce&redirect=https://github.com/browser-act/skills/tree/main, which would fetch and install external code as a required runtime dependency.