scrapfly-extraction

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

• Third-party content exposure detected (high risk: 0.90). This skill's SKILL.md explicitly shows scraping arbitrary public URLs (e.g., "Scrape + Extract in one flow" and "Inline extraction with the Scrape API" using client.scrape(ScrapeConfig(url=...))) and then feeding the scraped HTML/markdown/text into LLM extraction prompts or models, meaning untrusted third-party webpage content is read and interpreted and can materially influence actions, enabling indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill's runtime examples call client.scrape on https://web-scraping.dev/product/1 and then pass the fetched page content directly into the extraction/LLM flow, meaning remote page content can be injected into the model context (prompt-injection risk).