Generate usage statistics from your OpenAI Codex CLI data.

• Message Statistics: Total messages, daily/monthly breakdowns, top 10 busiest days
• Token Usage: Input, output, cached, and reasoning tokens per month
• Session Tracking: Number of sessions per month
• Copilot Pro Comparison: Compare your usage against the 1,500 messages/month limit

Choose either:

• Python: Python 3.6+ (no external dependencies)
• Go: Go 1.18+ (no external dependencies)

python3 generate-stats.py

# Build once
go build -o codex-usage-stats

# Run
./codex-usage-stats

Or run directly:

go run main.go

Output is written to ~/.codex/usage-statistics.md.

The script reads from:

1. Summary - Total messages, active days, date range, averages
2. Token Usage Summary - Total tokens broken down by type
3. Monthly Breakdown - Messages and sessions per month
4. Monthly Token Usage - Token consumption per month
5. Top 10 Busiest Days - Highest activity days
6. Daily Breakdown - All active days grouped by month
7. Usage Pattern Analysis - Copilot Pro quota comparison

## Token Usage Summary

| Metric | Tokens |
|--------|--------|
| **Total Tokens** | 4,369,677,393 (4369.7M) |
| **Input Tokens** | 4,345,071,360 (4345.1M) |
| **Output Tokens** | 24,334,033 (24.3M) |
| **Cached Input** | 4,197,451,520 (4197.5M) |
| **Reasoning Output** | 16,294,128 (16.3M) |

MIT