scraperapi-serp-intelligence

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

• Third-party content exposure detected (high risk: 0.85). The required workflow calls mcp__ScraperAPIRemote__google_search (and optionally mcp__ScraperAPIRemote__scrape on cited URLs) to ingest live Google SERP content and page text from third-party sites, which can include outsider-authored free text that becomes LLM-readable context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill issues runtime scrape calls to external pages (e.g., mcp__ScraperAPIRemote__scrape url="https://www.google.com/search?q=[url-encoded-query]" and cited-source URLs) and ingests that fetched HTML/text to directly drive the agent's outputs, so live external content controls the agent at runtime.