firecrawl-scraper

Warn

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • Unverifiable Dependencies (MEDIUM): The installation instructions npx skills add -g BenedictKing/firecrawl-scraper reference a GitHub repository belonging to an account ('BenedictKing') that is not in the trusted source list. This requires manual verification of the repository content before installation.
  • Indirect Prompt Injection (MEDIUM): The skill is designed to scrape web content and parse PDFs, which are external, untrusted data sources.
  • Ingestion points: External URLs and PDF files processed via the Firecrawl API.
  • Boundary markers: None identified in the skill metadata to delimit untrusted web content from agent instructions.
  • Capability inventory: Performs web crawling, content extraction, and PDF parsing.
  • Sanitization: No evidence of sanitization or filtering of the scraped content before it is returned to the agent context.
  • Data Exposure (INFO): The skill requires a Firecrawl API key. Users should ensure this is handled via environment variables as recommended in the best practices section to avoid credential leakage.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 16, 2026, 07:23 AM