firecrawl-research-index

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process content from external sources including PubMed, arXiv, and general web pages which are not under the agent's control.
  • Ingestion points: Data enters the agent context through paper abstracts via firecrawl_research_search_papers, full passages via firecrawl_research_read_paper, and web content via firecrawl_scrape.
  • Boundary markers: The instructions lack specific delimiters or instructions to treat retrieved research content as untrusted data, increasing the risk of the agent following instructions embedded within paper abstracts or scraped pages.
  • Capability inventory: The skill utilizes tools for network access (web search and scraping) and deep inspection of a research corpus.
  • Sanitization: There is no evidence of content sanitization or validation protocols for the external data being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 04:37 AM
Security Audit — agent-trust-hub — firecrawl-research-index