caveman-stats
juliusbrussee/caveman
Display real token usage and estimated savings for your Claude Code session.
What is caveman-stats?
Shows actual token consumption and cost savings from your current session by reading directly from the Claude Code session log. Triggers on /caveman-stats command with no AI estimation required.
- Display real token usage metrics from the current session
- Show estimated cost savings
- Read directly from Claude Code session logs without AI estimation
- Inject output via mode-tracker hook for immediate display
- Block further processing and return formatted statistics
How to install caveman-stats
npx skills add https://github.com/juliusbrussee/caveman --skill caveman-statsHow to use caveman-stats
- 1.Type /caveman-stats in your Claude Code session
- 2.The hook intercepts the command and reads the session log
- 3.Formatted statistics display immediately without model processing
- 4.Review token usage and estimated savings in the output
Use cases
- Monitor token consumption during a long coding session
- Verify actual API costs before they accumulate
- Track savings from using efficient coding practices
- Audit session efficiency at any point during development
- Compare token usage across different sessions or projects
- Developers using Claude Code
- Teams tracking API costs
- Users optimizing token efficiency
- Anyone monitoring session resource consumption
caveman-stats FAQ
No. The hook reads directly from the session log and returns the formatted stats, blocking further model processing.
It triggers when you type /caveman-stats in your Claude Code session.
Real token usage metrics and estimated cost savings for the current session.
This is actual data read directly from the Claude Code session log, not an AI estimation.
Full instructions (SKILL.md)
Source of truth, from juliusbrussee/caveman.
name: caveman-stats description: > Show real token usage and estimated savings for the current session. Reads directly from the Claude Code session log — no AI estimation. Triggers on /caveman-stats. Output is injected by the mode-tracker hook; the model itself does not compute the numbers.
This skill is delivered by hooks/caveman-stats.js (read by hooks/caveman-mode-tracker.js on /caveman-stats). The model does not need to do anything when this skill fires — the hook returns decision: "block" with the formatted stats as the reason. The user sees the numbers immediately.
Related skills
More from juliusbrussee/caveman and the wider catalog.
compress
Compress natural language memory files into caveman-speak to save input tokens while preserving code and structure.
cavecrew
Delegate to compressed subagents—investigator, builder, reviewer—to keep main context lean across long sessions.
caveman
Cut token usage ~75% with caveman-mode responses — full technical accuracy, zero fluff
caveman-commit
Ultra-compressed commit messages in Conventional Commits format—terse, exact, why-focused.
context-canary
Install a per-turn canary signal (e.g. starting every reply with the user's name and a turn counter) so silent context degradation becomes visible the moment it happens, and run a recovery protocol when the canary trips. Use when the user mentions a "canary", "context canary", or "canary check", asks to detect context rot / compaction / drift, says "you stopped using my name" or "did you lose context", asks "how degraded is your context", or wants an early-warning system for long agent sessions.
fuck-slop
>