#55th of 31 in Skills, plugins and rules

caveman

Proxy and skill that cut token use in coding agents

Stars
111k
License
Apache-2.0
Last commit
Oct 2026
Last release
Oct 2026
Language
Go

Overview

Caveman has two parts. A local proxy compresses what coding agents read (logs, CSV, YAML, JSON, test output, web pages), and a skill makes the agent reply in terse, answer-first text while code, paths and numbers stay verbatim. It works with Claude Code, Codex, Gemini CLI, Aider and 30+ other agents, and ships SDK middleware for the Vercel AI SDK, LangChain, OpenAI and Anthropic.

Who it is for: Heavy Claude Code or Codex users trying to cut token spend

Strengths

  • Proxy cut whole-session input tokens 33.2% in its own 54-run Claude Code benchmark
  • Skill-only install via `npx skills add`, no proxy needed
  • README lists losing cases: HTML got 9.9% worse, and the per-request billing caveat
  • TypeScript and Python middleware wraps existing LLM SDK calls

Weaknesses

  • No HTML compressor yet; the benchmark HTML session used more tokens
  • CLI sends usage telemetry by default; opt out with `caveman telemetry off`
  • Proxy needs Node.js 22.13+
  • Skill rules add about 1,160 tokens; cost impact with caching not measured

What it needs

  • no GPU
  • Docker + Compose
  • Needs Node.js 22.13+, Python 3.11+ (middleware only)
  • Models: Claude Code, Codex, Gemini CLI, Aider, Cursor

Also in Skills, plugins and rules

See all 31
Also in Skills, plugins and rules
RankProjectScore
1skillsAgent skills for planning, TDD, debugging and code review285.4k stars, MIT94 out of 100
2eccSkills, agents, hooks and rules that add a workflow to coding agents276.9k stars, MIT93 out of 100
3superpowersSkill library that enforces a plan-then-TDD workflow in coding agents297.4k stars, MIT89 out of 100
#4ponytailPrompt skill that makes coding agents write less, safer code160.8k stars, MIT89 out of 100
#6ui-ux-pro-max-skillDesign-system skill for AI coding assistants, with searchable UI styles134.7k stars, MIT88 out of 100