deep-research
Pass
Audited by Gen Agent Trust Hub on Aug 25, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted web data using search and scrape tools, which can contain adversarial instructions.
- Ingestion points: The firecrawl_scrape and crawling_exa tools used in SKILL.md ingest data from arbitrary URLs.
- Boundary markers: No delimiters or specific instructions are provided to help the agent distinguish between scraped data and its own system instructions.
- Capability inventory: The skill utilizes web searching, scraping, sub-agent task orchestration, and mentions saving reports to files.
- Sanitization: The skill lacks instructions for validating or sanitizing external content before it is synthesized into the final report.
Audit Metadata