firecrawl-knowledge-base

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted content from external websites, creating a surface for indirect prompt injection.
  • Ingestion points: Web content retrieved via Firecrawl's search and scrape tools as described in SKILL.md.
  • Boundary markers: The instructions lack explicit delimitation or instructions to ignore embedded commands within the scraped data.
  • Capability inventory: The agent is directed to write files to the local directory structure (e.g., .firecrawl/ folder) and perform network-based scraping.
  • Sanitization: No content sanitization or validation mechanisms are specified for the external data before it is formatted into the final deliverable.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:28 PM
Security Audit — agent-trust-hub — firecrawl-knowledge-base