The Agent Skills Directory

[COMMAND_EXECUTION]: The Python script scripts/firecrawl.py utilizes subprocess.run to spawn a worker instance of itself. This design is used to enforce strict HTTP operation deadlines and ensure process isolation for network requests. The command uses the current Python executable and the resolved path of the script itself with a hardcoded worker flag, which is a safe and robust implementation pattern.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process discovered URLs from external websites via the Frevana API. It proactively mitigates injection risks by including explicit instructions in SKILL.md for the agent to treat all collected content as untrusted source material rather than instructions. Data ingestion is limited to structured JSON results from a known vendor API.
[EXTERNAL_DOWNLOADS]: The skill performs legitimate network operations to the vendor's official API endpoint (ai-factory.frevana.com) to retrieve web discovery data. This communication is essential for the skill's primary function and uses standard HTTPS with bearer token authentication.

firecrawl-map