caveman-stats

Monitor token usage and savings in Claude Code sessions from logs.

1|Updated Jun 22, 2026
One-click install
npx skills add https://github.com/jchildree/The-QUESTION --skill caveman-stats-jchildree
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: caveman-stats
Source: https://github.com/jchildree/The-QUESTION/tree/main/.agents/skills/caveman-stats
Command: npx skills add https://github.com/jchildree/The-QUESTION --skill caveman-stats-jchildree

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill provides detailed insights into Claude Code session token usage, helping users understand and optimize their token consumption.

Core Features & Use Cases

  • Real-Time Usage Reporting: Monitors and reports actual input/output token usage for the current session.
  • Comparison Against Baseline: Estimates savings compared to a non-caveman baseline.
  • Session Log Analysis: Extracts statistics from the Claude Code session log without AI estimation.
  • Immediate Output: Integrates with the Claude Code interface to display formatted stats in real-time.

Quick Start

Run the '/caveman-stats' command to obtain the current session token usage and estimated savings.

Frequently Asked Questions about caveman-stats

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I track Claude Code session token usage?

You can track Claude Code session token usage by running a command that reads session logs to report exact input and output token counts without relying on AI estimation.

Can I monitor real-time token consumption in Claude Code?

Yes, you can monitor real-time token consumption in Claude Code by analyzing session logs, which provides immediate formatted stats directly within the interface.

How does session log analysis calculate token savings?

Session log analysis calculates token savings by comparing your exact input and output token counts against a non-caveman baseline usage metric.

Are Claude Code token stats reported with AI estimation?

No, Claude Code token stats are extracted directly from session logs to provide exact numbers, deliberately avoiding any AI estimation for precise usage tracking.

What is the best way to report input and output token counts for a coding session?

The best way to report input and output token counts for a coding session is to extract statistics directly from the session logs, ensuring precise real-time analysis and savings comparisons.