LINKED LIST [txt mode] ▸ GitHub - JuliusBrussee/caveman: 🪨 why...

home explore | log in

GitHub - JuliusBrussee/caveman: 🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.

github.com · first added by @hnl · 2026-09-25 · 2 upvotes

log in to save, upvote or flag this.

─── In 1 list ──────────────────────────────────────────

* Issue #789 by @hnl [code]

* GitHub - zachahn/vomit: Clean up Claude's token vomit with a separate LLM. Save your tokens, Opus is hopeless (from the discussion)

─── Discussions ────────────────────────────────────────

* Caveman: Why use many token when few token do trick

904 pts · 366 comments · node

─── From the discussion ────────────────────────────────

* The Grug Brained Developer

grugbrain.dev · node

* The Benefits of a Concise Chain of Thought on Problem-Solving in Large Language Models

arxiv.org · node

* https://arxiv.org/pdf/2604.00025

arxiv.org · node

* Let's Think Dot by Dot: Hidden Computation in Transformer Language Models

arxiv.org · node

* Extend Claude with skills - Claude Code Docs

code.claude.com · node

─── Also saved alongside this ──────────────────────────

* Music for Programming

3 upvotes · 2026-01-31

* Sam Altman May Control Our Future—Can He Be Trusted?

2 upvotes · 2026-09-25

* Project Glasswing: Securing critical software for the AI era

2 upvotes · 2026-09-25

* GitHub - arman-bd/guppylm: A ~9M parameter LLM that talks like a small fish.

2 upvotes · 2026-09-25