firecrawl
Pass
Audited by Gen Agent Trust Hub on Jun 26, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the official
firecrawl-clipackage (v1.16.2) via NPM. It also manages the installation of related agent skills and workflows through the CLI's setup commands. - [COMMAND_EXECUTION]: Web interaction is performed by executing
firecrawlCLI commands in a shell environment. The instructions require quoting URLs to mitigate shell injection risks. - [DATA_EXFILTRATION]: Data is sent to the official Firecrawl service for processing and feedback. This behavior is transparent, documented, and essential for the tool's core functionality.
- [PROMPT_INJECTION]: The skill identifies the risk of indirect prompt injection from scraped web data and prescribes safe handling techniques, including saving results to local files and using incremental reading tools like
grepto avoid context contamination.
Audit Metadata