|
A Claude Code skill that makes Claude talk like a caveman — cutting ~75% of tokens while keeping full technical accuracy. Save money. Go faster. Same result.
claude install-skill JuliusBrussee/caveman
useMemo."< not <=. Fix:"Same fix. 75% less word. Brain still big.
Articles (a, an, the), pleasantries, and hedging — all gone. "I'd be happy to help" dies here.
Technical terms, error messages, code blocks, and git commits always written normally. Caveman not stupid.
Fewer tokens to generate means faster responses. Less waiting. More shipping.
Claude API billed per token. Cut 75% output tokens = cut 75% output cost. Math easy.
Type "/caveman" to activate. "normal mode" to stop. Switch mid-session without losing context.
One command. Zero config. Works immediately with any Claude Code project.
Normal LLM wastes tokens on garbage. "I'd be happy to help you with that" — 8 wasted tokens. "The reason this is happening is because" — 7 wasted tokens. Caveman say what need saying. Then stop.
Learn More →One command. Instant results. Zero config.