seo-firecrawl

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external websites via crawling and scraping tools, which can be exploited if a processed website contains malicious instructions targeting the AI agent.\n
  • Ingestion points: External web content retrieved via firecrawl_crawl and firecrawl_scrape (SKILL.md).\n
  • Boundary markers: Absent. The skill does not include instructions to use delimiters or to ignore instructions embedded within the scraped data.\n
  • Capability inventory: The skill performs network operations via MCP tools and writes analysis data to the local filesystem in the .seo-cache/ directory.\n
  • Sanitization: Absent. There is no mention of sanitizing or filtering the scraped markdown or HTML content before processing.\n- [COMMAND_EXECUTION]: The documentation instructs the user or agent to execute a shell script for installation and setup of the Firecrawl extension.\n
  • Evidence: Instruction to run the local script ./extensions/firecrawl/install.sh in SKILL.md.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 07:39 AM
Security Audit — agent-trust-hub — seo-firecrawl