The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from the web via the Firecrawl API (search, scrape, extract, and Alexandria tools).
Ingestion points: Web content retrieved in scripts/firecrawl.py through the search(), scrape(), extract(), and run() functions.
Boundary markers: The script outputs raw Markdown or JSON content from the API without explicit delimiters or instructions to the agent to ignore embedded commands.
Capability inventory: The CLI script allows writing output to local files via the -o/--output argument.
Sanitization: No specific sanitization or filtering of the retrieved web content is performed before it is presented to the agent.
[COMMAND_EXECUTION]: The skill uses a subprocess call to manage credentials securely.
Evidence: scripts/firecrawl.py calls subprocess.run() to execute the sops binary for decrypting the FIRECRAWL_API_KEY. This is a legitimate use case for secret management and the arguments are strictly controlled by the script logic.

firecrawl-api