The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses standard curl commands to interact with the Firecrawl API and kubectl to retrieve application logs. These commands are localized to the triage and evidence collection steps and are appropriate for the stated incident response use case.
[DATA_EXPOSURE]: Authentication is handled via the $FIRECRAWL_API_KEY environment variable, avoiding hardcoded secrets. The runbook explicitly instructs users/agents to preserve "redacted evidence only" and keep sensitive data in approved stores.
[INDIRECT_PROMPT_INJECTION]: The skill involves processing logs and API responses. While this presents a theoretical surface for injection, the skill includes explicit instructions for data redaction and safe evidence handling, which mitigates the risk of processing untrusted content.

firecrawl-incident-runbook