firecrawl

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill retrieves and processes arbitrary content from the web, which creates an attack surface for indirect prompt injection attacks where malicious data can influence the agent's behavior.
  • Ingestion points: The scrape, crawl, and search functions in scripts/firecrawl.sh fetch external web content and pipe it to standard output.
  • Boundary markers: Absent. The skill does not implement delimiters or warnings to separate external data from system instructions.
  • Capability inventory: The skill possesses the ability to execute shell commands (Bash), perform network requests (curl), and read credentials via pass.
  • Sanitization: Absent. The content retrieved from external URLs is returned directly to the agent context without sanitization or filtering.
  • [EXTERNAL_DOWNLOADS]: The script performs network operations targeting api.firecrawl.dev. This is recognized as a well-known service for automated web data extraction.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 10:56 AM