skills vs caveman
Two of the top skills, plugins and rules, side by side: score, setup, license, activity and what each review found.
skills
Agent skills for planning, TDD, debugging and code review
caveman
Proxy and skill that cut token use in coding agents
| What we compare | skills | caveman |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 98, widely used | 79, popular |
| Freshness | 100, active | 100, active |
| Maintenance | 93, healthy | 97, healthy |
| Easy to run | Not scored for this kind | Not scored for this kind |
| Agent-ready | 70, partly | 70, partly |
| Facts from GitHub and the README | ||
| Stars | 287.7k | 111.2k |
| License | MIT (permissive) | Apache-2.0 (permissive) |
| Last commit | Oct 2026 | Oct 2026 |
| Last release | Oct 2026 | Oct 2026 |
| Language | Shell | Go |
| Docker | No | Yes |
| GPU | Not needed | Not needed |
| arm64 or Apple Silicon | Not stated | Not stated |
skills
A collection of Markdown skill files for coding agents, split into engineering and productivity groups. It covers grilling sessions that build a project glossary and ADRs, spec and ticket generation, red-green-refactor TDD, a bug diagnosis loop, and diff review. Installs as a plugin for Claude Code, Codex and GitHub Copilot, via gemini skills install for Gemini CLI, or as editable files through npx skills.
Who it is for: Engineers using Claude Code, Codex, Copilot or Gemini CLI on real codebases
Strengths
- Small, composable skills that can be edited; README says they work with any model
- Plugin install for Claude Code, Codex and Copilot, with automatic updates
- Setup skill configures issue tracker (GitHub, GitLab, local files), triage labels and doc locations
- Separates user-invoked orchestration skills from model-invoked discipline skills
Weaknesses
- Manual updates for Gemini CLI and npx installs
- Installing both plugin and npx copies gives every skill twice
- Must run /setup-matt-pocock-skills once per repo before the engineering skills
- Architecture skill is a survey only and will not untangle an old codebase
- no GPU
- Needs Issue tracker (GitHub, GitLab or local files)
- Models: Any model (per README)
caveman
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
- no GPU
- Docker + Compose
- Needs Node.js 22.13+, Python 3.11+ (middleware only)
- Models: Claude Code, Codex, Gemini CLI, Aider, Cursor
More in Skills, plugins and rules
Building with these?
We help teams pick an open-source AI stack and run it in production. Client work never changes a rank.