firecrawl-search

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill utilizes npx firecrawl to run the Firecrawl CLI, which involves downloading the package from the NPM registry. This is the standard distribution method for the author's tools.\n- [COMMAND_EXECUTION]: The skill executes firecrawl and npx firecrawl commands via the Bash tool to perform web searches and scrape page content.\n- [DATA_EXFILTRATION]: The firecrawl search-feedback utility sends telemetry data (ratings and missing content information) to the vendor's API. This is used for search optimization and credit refunds; the skill documentation includes instructions on how to disable this via environment variables.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes content from the public web.\n
  • Ingestion points: Web search results and full-page markdown content retrieved via the firecrawl search command.\n
  • Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded in the external content.\n
  • Capability inventory: The skill has access to the Bash tool, allowing the execution of CLI utilities like firecrawl and jq.\n
  • Sanitization: No content filtering or sanitization is mentioned for the scraped web data.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 04:39 AM
Security Audit — agent-trust-hub — firecrawl-search