firecrawl-agent
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
firecrawlCLI andnpx firecrawlto perform data extraction tasks. These commands are restricted to the vendor's official tooling. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to navigate and ingest content from external websites, which presents a potential surface for indirect prompt injection if the processed web content contains instructions intended to manipulate the agent.
- Ingestion points: External website content accessed during the autonomous extraction process.
- Boundary markers: Not explicitly defined in the skill instructions; relies on the underlying agent model's safety constraints.
- Capability inventory: Includes file system writing (via the
-ooutput flag) and command execution for the Firecrawl CLI. - Sanitization: Not explicitly defined in the provided markdown; extraction is handled by the Firecrawl service.
Audit Metadata