caveman-stats

An agent skill by JuliusBrussee, from JuliusBrussee/caveman. Tags: analytics, cli, local-workflow, tracking.

What it does

Show recorded output and cache-read token usage and mode attribution for the current Claude Code session, or locate the host's native usage report. Trigger: /caveman-stats.

Install

With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:

npx skills add JuliusBrussee/caveman --skill caveman-stats

Or copy the skill folder into Claude Code's skills directory by hand (~/.claude/skills for every project, or .claude/skills inside one):

git clone --depth 1 https://github.com/JuliusBrussee/caveman
cp -r caveman/plugins/caveman/skills/caveman-stats ~/.claude/skills/caveman-stats

Safety box score

Grade A, minimal risk: 0 of 100 risk points, rated 2026-10-16. The grade rates what the skill and its scripts can reach on the machine of whoever installs it, across eight categories from shell execution to secrets access. How the score works.

This is a read-only reporting skill: it prints Claude Code usage data supplied by an existing hook and, in Gemini, points users to built-in stats commands. It does not itself run shell commands, write files, or access secrets; the main thing to verify is that the referenced hook scripts stay local and only report usage.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	`SKILL.md` only instructs the agent to print hook-provided output verbatim; it contains no shell commands, code evaluation, or script execution steps.
Network access	0 (none)	The skill mentions local session statistics and native usage reports only; it does not download from or send data to remote hosts.
Filesystem reach	0 (none)	No file creation, deletion, or paths outside the project are requested; the skill just reads and displays usage information.
Secrets & credentials	0 (none)	Although it talks about token usage, it does not ask for API keys, cookies, passwords, `.env` files, or browser sessions.
Privilege & persistence	0 (none)	There is no sudo, global install, service registration, or shell/OS configuration in the skill text.
Instruction hijack surface	0 (none)	The text does not tell the agent to ignore its rules, hide actions, or follow remote instructions; it only distinguishes host-specific behavior.
Irreversible actions	0 (none)	The skill is informational only; it does not post, deploy, delete, pay, or otherwise take external actions.
Transparency	0 (none)	The reviewed file is plain text markdown; there are no binaries, base64 blobs, or uninspectable installers.

Before installing

Inspect `src/hooks/caveman-stats.js` and `src/hooks/caveman-mode-tracker.js` to confirm they only read local stats and do not transmit data.
Verify any Gemini `/stats` guidance is just a read-only built-in command path, not a custom prompt that executes other actions.
Review `docs/HONEST-NUMBERS.md` if you need to confirm how the reported counts are derived and whether any historical fields are only displayed.

Rated 2026-10-16 by an automated review of 1 file from the skill's repository.

Source

Repository: JuliusBrussee/caveman (all skills from this repository)
Path: plugins/caveman/skills/caveman-stats/SKILL.md
Branch: main
Collection: caveman
Updated: 2026-10-05

br — Official skill for beads_rust (`br`), a local-first, dependency-aware issue tracker for AI agents.
caveman — Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens".
megacave — Caveman in Classical Chinese: 文言文 register, far fewer characters, technical terms verbatim. Invoke only with /megacave or /caveman wenyan.
ultracave — Caveman at maximum compression: fragments, one word when one word is enough, each fact once. Invoke only with /ultracave or /caveman ultra.
agents-get-started — Use when a developer wants to create a new agent project or get started with AgentCore.
analytics — When the user wants to set up, improve, or audit analytics tracking and measurement.