firecrawl

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The Bash script scripts/firecrawl.sh orchestrates API calls and data processing using standard tools like curl and jq.
  • [EXTERNAL_DOWNLOADS]: The skill fetches content from the internet via api.firecrawl.dev, the official endpoint for Firecrawl, which is a recognized service provider for web scraping.
  • [PROMPT_INJECTION]: The skill has an attack surface for Indirect Prompt Injection because it processes untrusted data from external websites.
  • Ingestion points: Data enters the agent's context through the scrape, crawl, and search functions defined in scripts/firecrawl.sh.
  • Boundary markers: The script does not implement specific delimiters or instructions to ignore potential commands embedded in the scraped text.
  • Capability inventory: The script has the ability to make network requests (curl) and retrieve secrets from the user's password manager (pass).
  • Sanitization: The output is structured using jq, but the textual content extracted from the web is not sanitized for malicious instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 06:59 AM