The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the mushi CLI and supabase CLI to perform diagnostic checks, including mushi doctor, mushi status, and mushi deploy check. These tools are associated with the skill's vendor (kensaurus) and represent standard operational functionality for a health check utility.
[DATA_EXPOSURE]: The skill provides instructions for managing 'BYOK' (Bring Your Own Key) keys via the mushi keys command and MCP tools. While this involves handling API keys for services like Anthropic and Firecrawl, the operations are performed locally or directed to the user's configured project environment.
[PROMPT_INJECTION]: The skill processes data from edge function logs and database queries (e.g., qa_story_runs). This constitutes an indirect prompt injection surface where content in logs could theoretically influence agent behavior. However, this is inherent to the diagnostic purpose of the skill and no exploitable injection content was detected.

mushi-health