firecrawl-web
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION] (LOW): Indirect Prompt Injection risk. * Ingestion points: Scraped markdown and search results are retrieved from external URLs in 'fc.py' functions 'scrape_markdown', 'search_web', and 'crawl_docs'. * Boundary markers: Absent; the scraped content is returned directly to the agent context without delimiters or warnings. * Capability inventory: The skill has 'Bash', 'Read', and 'Write' permissions; 'fc.py' can write files to the local disk. * Sanitization: Absent; content from the web is not escaped or validated before being passed to the agent.
- [EXTERNAL_DOWNLOADS] (LOW): The 'fc.py' script uses 'urllib.request.urlretrieve' to download screenshot files from remote URLs returned by the Firecrawl API.
Audit Metadata