deep-research
Pass
Audited by Gen Agent Trust Hub on Apr 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to scrape and synthesize information from external websites, creating a surface for potential indirect prompt injection attacks where malicious instructions on web pages could influence the agent's behavior.
- Ingestion points: Content is retrieved from the internet using firecrawl_scrape and crawling_exa tools.
- Boundary markers: There are no explicit markers or safety instructions used to isolate scraped content from the agent's core instructions.
- Capability inventory: The skill has search, scrape, and parallel sub-agent execution capabilities.
- Sanitization: No explicit logic is provided to sanitize or validate external content before processing.
- [SAFE]: The skill uses established research tools and follows a transparent methodology without any detected malicious code, credential harvesting, or unauthorized persistence mechanisms.
Audit Metadata