Analyze Prompt Command

Use the prompt-analyzer agent to evaluate the user's prompt and provide a quality score.

Process

1. Get the prompt:

   • If arguments provided: {{$ARGUMENTS}}
   • If a file path is given: Read the file using the Read tool
   • If no prompt available: The agent will report an error

2. Invoke the agent: Use the Task tool with subagent_type: "prompt-analyzer" and pass:

   • The prompt to analyze
   • Target model (if known)

3. Present results: The agent returns a detailed analysis report including:

   • Quality score (0-100)
   • Breakdown by category (Clarity, Context, Structure, Output Spec, Techniques, Model Fit)
   • Strengths and issues with severity ratings
   • Suggestions for improvement (without implementing them)