caveman-help
An agent skill by JuliusBrussee, from JuliusBrussee/caveman. Tags: cli, documentation, help, local-workflow.
What it does
Quick-reference card for the three caveman skills and their commands. Trigger: /caveman-help or "caveman help".
Install
With the skills CLI, which installs into Claude Code, Codex, Cursor and other agents:
npx skills add JuliusBrussee/caveman --skill caveman-helpOr 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/skills/caveman-help ~/.claude/skills/caveman-helpSafety 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 skill is a read-only reference card: it explains caveman modes, triggers, and configuration, but it does not ship executable code or any action commands. The main thing to notice is that it points to a home-directory config path and external docs, but the skill itself does not modify files or contact services.
| Category | Level | Why |
|---|---|---|
| Code & shell execution | 0 (none) | No shell commands, code evaluation, or native scripts are included; both files are plain markdown reference text. |
| Network access | 0 (none) | The only remote reference is a documentation link (`https://github.com/JuliusBrussee/caveman`); the skill itself does not fetch or send data. |
| Filesystem reach | 0 (none) | It describes a config file path (`~/.config/caveman/config.json`) but does not read, write, or delete any files itself. |
| Secrets & credentials | 0 (none) | No tokens, keys, passwords, or login flows are requested or handled. |
| Privilege & persistence | 0 (none) | There is no sudo, service install, shell startup edit, or other privilege-changing behavior. |
| Instruction hijack surface | 0 (none) | The text gives normal formatting and mode instructions, but nothing that tells the agent to skip confirmations, hide behavior, or follow untrusted remote instructions. |
| Irreversible actions | 0 (none) | It is informational only; it does not post, deploy, delete, or otherwise act on the user's behalf. |
| Transparency | 0 (none) | All shipped content is readable markdown; there are no binaries, blobs, or generated installers. |
Before installing
- Read `SKILL.md` to confirm it is documentation-only and does not ship scripts.
- If you use caveman tooling, inspect whether it actually writes `~/.config/caveman/config.json` before enabling it.
- Treat the GitHub link as optional reference, not something the skill itself will fetch.
Rated 2026-10-16 by an automated review of 2 files from the skill's repository.
Source
- Repository
- JuliusBrussee/caveman (all skills from this repository)
- Path
- skills/caveman-help/SKILL.md
- Branch
- main
- Updated
- 2026-10-05
Related skills
- caveman — Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens".
- caveman-stats — 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.
- 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.
- bambu-labs — Dry-run, upload, and cautiously initiate local Bambu Lab print jobs from validated plain `.gcode`, using Bambu LAN FTPS/MQTT handoffs.