The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to navigate and extract data from external websites. This involves processing untrusted third-party content, which presents a surface for indirect prompt injection where malicious instructions embedded in a website could attempt to influence the agent's behavior.
Ingestion points: Data is ingested from external websites via the firecrawl agent command (SKILL.md).
Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions for the content being processed.
Capability inventory: The skill has access to the filesystem (via the -o flag for JSON output), network access (for scraping), and shell command execution (SKILL.md).
Sanitization: There is no mention of sanitization or filtering of the content extracted from websites before it is processed by the agent.
[COMMAND_EXECUTION]: The skill uses shell commands through the firecrawl CLI and npx firecrawl-cli to perform its tasks, including writing output to the .firecrawl/ directory.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx to execute the firecrawl-cli package, which involves downloading the tool from the npm registry if it is not already available in the environment.

firecrawl-agent