caveman-stats

Analyze Claude Code session JSONL logs for token usage and savings.

2|1|Updated Sep 25, 2025
One-click install
npx skills add https://github.com/istinataFTS/LiftLeagueLegends --skill caveman-stats-istinatafts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: caveman-stats
Source: https://github.com/istinataFTS/LiftLeagueLegends/tree/main/fitness_tracker/.agents/skills/caveman-stats
Command: npx skills add https://github.com/istinataFTS/LiftLeagueLegends --skill caveman-stats-istinatafts

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides accurate token usage and savings data for Claude Code sessions, offering transparency into session costs and efficiency improvements.

Core Features & Use Cases

  • Token Usage Reporting: Showcases the exact token usage for input and output.
  • Savings Estimation: Compares the current usage with a baseline to estimate potential savings.
  • Session Tracking: Integrates with Claude Code sessions for on-the-fly analytics.
  • Use Case: After running a session, use the caveman-stats command to understand how many tokens were used and compare them to the baseline for savings.

Quick Start

Trigger caveman-stats in a Claude Code session to see the session's token usage and 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 token usage in Claude Code sessions?

To track token usage in Claude Code sessions, trigger the caveman-stats command to analyze JSONL session logs for actual input and output token counts. It provides on-the-fly analytics and transparency into session costs.

How does token savings estimation work in Claude Code?

Token savings estimation works by comparing your current Claude Code session token usage against an estimated baseline. It analyzes JSONL logs to calculate the difference, helping you evaluate efficiency improvements and potential savings.

What are JSONL session logs and do I need them for token tracking?

JSONL session logs are the data files generated by Claude Code that record session activity. You need these logs available for the token tracking command to analyze actual token counts and evaluate efficiency against a baseline.

Can I see real-time token usage while running a Claude Code session?

Yes, you can see token usage analytics for your Claude Code session. After running a session, trigger the caveman-stats command to integrate with the session logs and report exact input and output token counts.

What is the best way to evaluate Claude Code token efficiency?

The best way to evaluate Claude Code token efficiency is to compare actual token usage from JSONL logs against an estimated baseline. This comparison reveals potential savings and highlights how efficiently tokens are being used.

Does caveman-stats require any external dependencies or components?

No, caveman-stats does not require any external dependencies or components. It operates independently to analyze Claude Code JSONL session logs and report token usage and savings data without additional setup.