---
name: Caveman Stats
slug: caveman-stats
category: Data
description: Caveman Stats shows real token usage and estimated savings for the current Claude Code session by reading the session log directly. Use /caveman-stats to see net savings after rule overhead, without AI estimation.
github: "https://github.com/JuliusBrussee/caveman/tree/main/skills/caveman-stats"
language: Go
stars: 97841
forks: 5640
install: "npx degit https://github.com/JuliusBrussee/caveman/tree/main/skills/caveman-stats ~/.claude/skills/caveman-stats"
installs_to: ~/.claude/skills/caveman-stats
source_path: skills/caveman-stats/SKILL.md
collection_size: 20
category_size: 668
collection_url: "https://dirskills.com/collections/JuliusBrussee/caveman"
added: 2026-08-13T07:35:51.923Z
last_synced: 2026-08-13T07:35:51.923Z
canonical_url: "https://dirskills.com/skills/caveman-stats"
---

# Caveman Stats

Caveman Stats shows real token usage and estimated savings for the current Claude Code session by reading the session log directly. Use /caveman-stats to see net savings after rule overhead, without AI estimation.

**Install:**

```bash
npx degit https://github.com/JuliusBrussee/caveman/tree/main/skills/caveman-stats ~/.claude/skills/caveman-stats
```

## README

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.

Output also includes `Est. rule overhead` and `Est. net` lines wherever a savings estimate exists with a known turn count. Rule overhead is the estimated per-turn INPUT-token cost of the injected caveman rules (default 1,250 tokens/turn, override with `CAVEMAN_RULE_OVERHEAD_TOKENS`) times the turn count. Net is savings minus that overhead — when negative, the output says so plainly and suggests turning caveman off for that workload, rather than hiding the net-negative regime behind a gross-savings number (see `docs/HONEST-NUMBERS.md`).