megacave

An agent skill by JuliusBrussee, from JuliusBrussee/caveman. Tags: cli, copy-editing, local-workflow, writing.

What it does

Caveman in Classical Chinese: 文言文 register, far fewer characters, technical terms verbatim. Invoke only with /megacave or /caveman wenyan. Stays on until "stop caveman" or "normal mode".

Install

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

npx skills add JuliusBrussee/caveman --skill megacave

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/megacave ~/.claude/skills/megacave

Safety box score

Grade A, minimal risk: 5 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 pure response-style modifier: it forces the assistant into persistent Classical Chinese until a reset phrase, while preserving code, commands, and identifiers verbatim. It does not ship scripts, reach the network, or touch files; the main risk is behavioral injection/persistence rather than machine or data access.

Category levels, from 0 (none) to 3 (high)
Category	Level	Why
Code & shell execution	0 (none)	No scripts, shell commands, or code evaluation are shipped; `SKILL.md` is text-only.
Network access	0 (none)	No downloading, posting, or remote data transfer is described anywhere in `SKILL.md`.
Filesystem reach	0 (none)	The skill does not read, write, or delete files; it only changes how responses are phrased.
Secrets & credentials	0 (none)	No credentials, tokens, `.env` files, keychains, or browser sessions are mentioned.
Privilege & persistence	0 (none)	No sudo, installs, services, shell startup edits, or OS/git settings are requested.
Instruction hijack surface	1 (low)	It imposes a persistent persona/style change — `Every response, whole session, until "stop caveman" or "normal mode"` — which is a model-behavior override, but it does not instruct bypassing safeguards or hiding content.
Irreversible actions	0 (none)	The skill does not send messages, deploy, publish, pay, book, or delete anything.
Transparency	0 (none)	There are no binaries, generated blobs, or unreadable installers; the entire skill is plain markdown.

Before installing

Confirm you want a persistent style/persona change for the whole chat session.
Verify there are no accompanying scripts or hooks in the repo beyond `SKILL.md`.
If you use it, test the reset path (`stop caveman` / `normal mode`) in a scratch conversation first.

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/megacave/SKILL.md
Branch: main
Collection: caveman
Updated: 2026-10-05

caveman — Terse caveman voice: answer first, fluff gone, every technical fact kept. Use for /caveman, "caveman mode", "talk like caveman", "be brief", "less tokens".
ultracave — Caveman at maximum compression: fragments, one word when one word is enough, each fact once. Invoke only with /ultracave or /caveman ultra.
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.
agents-get-started — Use when a developer wants to create a new agent project or get started with AgentCore.
copy-editing — When the user wants to edit, review, or improve existing marketing copy, or refresh outdated content.
google-docs — Author and edit Google Docs with the google-docs tools - structured writing, character and paragraph styling, lists, tables, images, headers/footers.