The Agent Skills Directory

[COMMAND_EXECUTION]: The skill operates by executing the firecrawl binary via the shell. It also provides examples for using shell utilities such as jq, grep, head, wc, and xargs to process scraped data and manage parallel execution.
[EXTERNAL_DOWNLOADS]: The documentation instructs the user to install the firecrawl-cli package globally via npm install -g firecrawl-cli, which is a standard method for installing this specific utility.
[PROMPT_INJECTION]: The skill facilitates the ingestion of third-party web content, creating a vulnerability to indirect prompt injection where an attacker-controlled website could include instructions aimed at manipulating the agent's behavior.
Ingestion points: Data is pulled into the agent's context through the scrape, search, and crawl subcommands as described in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided to the agent when reading the generated markdown files.
Capability inventory: The skill has the ability to execute shell commands, write to the local file system (specifically the .firecrawl/ directory), and perform network requests via the CLI.
Sanitization: The skill relies on Firecrawl's internal processing to return 'clean markdown', but it does not implement its own sanitization to filter out potentially malicious natural language instructions within that content.

firecrawl-cli