The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill allows the agent to navigate websites and extract data autonomously. This exposes the agent to untrusted data from the open web which could contain malicious instructions designed to subvert the agent's behavior.
Ingestion points: The firecrawl agent command processes arbitrary web content from external URLs.
Boundary markers: The skill does not define specific boundaries or instructions to ignore embedded commands within the extracted web content.
Capability inventory: The agent has access to Bash and can write output to the filesystem (e.g., the .firecrawl/ directory).
Sanitization: There is no explicit sanitization step for the scraped content before the agent processes it.
[COMMAND_EXECUTION]: The skill relies on executing firecrawl and npx firecrawl-cli via the Bash tool to perform its primary functions.
[EXTERNAL_DOWNLOADS]: The tool npx firecrawl-cli is listed in allowed-tools. npx fetches and executes the CLI from a registry at runtime. This is standard functionality for the vendor's provided tooling.

firecrawl-agent