firecrawl-instruct

Pass

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx firecrawl to download and run the Firecrawl CLI from the public npm registry.
  • [COMMAND_EXECUTION]: Instructs users on how to execute browser automation tasks and arbitrary code snippets in Python, Node.js, or Bash via the firecrawl interact --code command for precise page control.
  • [PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection because it processes data from untrusted external web pages that could contain instructions designed to manipulate the browser session.
  • Ingestion points: Arbitrary URLs provided to the firecrawl scrape command in SKILL.md.
  • Boundary markers: None identified in the instructional prompts.
  • Capability inventory: Capabilities include clicking elements, filling forms, and executing code via the firecrawl interact command in SKILL.md.
  • Sanitization: No explicit sanitization or filtering of the scraped web content is mentioned in the instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 15, 2026, 08:19 PM