The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill utilizes npx firecrawl, which facilitates the download and execution of the Firecrawl package from the npm registry. This is a standard deployment method for this tool.
[COMMAND_EXECUTION]: The skill executes shell commands via the Bash tool to run the scraper. It passes user-supplied URLs and query strings directly as command-line arguments.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection by processing arbitrary content from external URLs.
Ingestion points: Untrusted data enters the agent context via webpage content fetched from URLs specified in the commands.
Boundary markers: No specific delimiters or instructions are provided to help the agent distinguish between its core instructions and the potentially untrusted content fetched from the web.
Capability inventory: The agent has access to the Bash tool for executing commands and writing output to the local file system.
Sanitization: The skill lacks explicit sanitization or filtering of the retrieved markdown content, which could allow malicious instructions embedded in a webpage to influence the agent's behavior.

firecrawl-scrape