firecrawl-instruct
Pass
Audited by Gen Agent Trust Hub on Apr 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes
firecrawlandnpx firecrawlthrough theBashtool to perform browser automation tasks, including scraping, clicking elements, and filling forms in a live session. - [EXTERNAL_DOWNLOADS]: The use of
npx firecrawlinvolves fetching thefirecrawlpackage from the official NPM registry. This is a standard distribution method for the author's tooling. - [PROMPT_INJECTION]: The skill is subject to indirect prompt injection because it interacts with and processes untrusted data from arbitrary web pages.
- Ingestion points: External web content is ingested into the agent context via the
firecrawl scrapecommand. - Boundary markers: The skill lacks explicit instructions for the agent to distinguish between its own objectives and potentially malicious instructions embedded in the scraped HTML or JavaScript.
- Capability inventory: The skill possesses capabilities for network interaction, browser state persistence (via profiles), and data extraction.
- Sanitization: No specific sanitization or filtering of the external page content is defined before the agent performs interactions based on that content.
Audit Metadata