How to Keep Using diagnose-mac After Gemini Pro Hits Its Limit

In the last post, I shared my diagnose-mac function that dumps CPU, memory, thermal, and powermetrics output into Gemini CLI so it can tell me which process is cooking my laptop. It worked great until I hit my gemini-3.1-pro-preview limits for the day. Then the whole flow broke.

I wanted to keep the exact same one-command workflow even when Pro was exhausted. So I updated the script to support both Pro and Flash models, with a simple flag switch and an optional clipboard mode.

The Updated `diagnose-mac` Function

The new version keeps three ideas in mind:

Default to Pro. Use gemini-3.1-pro-preview when there is quota.
Fall back fast. Use --flash to switch to gemini-3.1-flash-preview when Pro is tapped out.
Optional clipboard. Use -c or --copy-data to just capture the full diagnostics blob into the clipboard if you want to paste it into another tool.

~/.zshrc

# Load shared shell functions
source ~/.zsh_functions

~/.zsh_functions

# diagnose-mac history
# 2026-03-09: v1, Pro-only version with optional --copy-data flag
# 2026-03-13: v2, add --flash / --pro model switches and keep clipboard flag

diagnose-mac () {
  local copy_to_clipboard=false 
  local model="gemini-3.1-pro-preview" # Default model

  # Parse arguments
  while [[ "$#" -gt 0 ]]; do
    case "$1" in
      -c | --copy-data) 
        copy_to_clipboard=true 
        shift 
        ;;
      --flash) 
        model="gemini-3.1-flash-preview" 
        shift 
        ;;
      --pro) 
        model="gemini-3.1-pro-preview" 
        shift 
        ;;
      *) 
        echo "Unknown option: $1"
        return 1
        ;;
    esac
  done

  echo "Collecting diagnostic data (requires sudo for powermetrics)..."
  echo "Using model: $model"

  local diagnostics
  diagnostics=$(
    echo -e "Please figure out what's wrong with my computer. Why is it running hot and/or slow today, and why are the fans acting up? Are there any CPU or RAM hogs?\n\n"
    echo -e "<output cmd="top_cpu">\n$(top -l 2 -o cpu -n 15)\n</output>\n"
    echo -e "<output cmd="top_mem">\n$(top -l 1 -o mem -n 15)\n</output>\n"
    echo -e "<output cmd="pmset">\n$(pmset -g therm)\n</output>\n"
    echo -e "<output cmd="powermetrics">\n$(sudo powermetrics -n 1)\n</output>"
  ) 

  if [ "$copy_to_clipboard" = true ]; then
    echo "$diagnostics" | pbcopy
    echo "Data saved to clipboard."
  fi

  cd /tmp && echo "$diagnostics" | gemini -m "$model" -p "Analyze these diagnostics for performance issues using your deep reasoning/thinking capabilities. Identify the specific PID or process causing issues."
  
  echo -e "\nAnalysis complete."
}

How I Actually Use It

On a normal day, I just run:

diagnose-mac

The script gathers all the metrics, sends them to Gemini Pro, and tells me exactly which PID is misbehaving. If I see Pro hitting its daily limit, I can immediately fall back to Flash without changing anything else in my workflow:

diagnose-mac --flash

And if Gemini CLI itself is being flaky or I want to paste the blob into another tool (or a web interface), I can run:

diagnose-mac --copy-data

That copies the entire diagnostics payload to the clipboard so I can drop it straight into a chat window without rerunning top or powermetrics.

Enjoyed this post?

Get notified when I publish something new. No spam, unsubscribe anytime.

← How to Diagnose a Slow Mac with Gemini CLI Screen & Webcam Daemons →

The Updated diagnose-mac Function

How I Actually Use It

The Updated `diagnose-mac` Function