scraperapi-research-agent
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches search results and webpage content from ScraperAPI's official API domain (
api.scraperapi.com). This is the core intended functionality and uses standard HTTPS requests. - [PROMPT_INJECTION]: As the skill ingests untrusted content from the web to generate reports, it has an inherent surface for indirect prompt injection.
- Ingestion points: Web content is fetched in
scripts/research_agent.pythrough thescrape_pagefunction. - Boundary markers: Scraped content is uploaded as separate document artifacts using the Anthropic Files API, which provides a structural boundary from the core system prompt.
- Capability inventory: The skill synthesizes text and can write the final report to a local file, but it does not perform any shell command execution or dynamic code evaluation based on the scraped content.
- Sanitization: Content is truncated to 20,000 characters and processed through ScraperAPI's markdown conversion service, reducing the risk of malicious payload injection.
Audit Metadata