ecc vs caveman
Two of the top skills, plugins and rules, side by side: score, setup, license, activity and what each review found.
ecc
Skills, agents, hooks and rules that add a workflow to coding agents
caveman
Proxy and skill that cut token use in coding agents
| What we compare | ecc | caveman |
|---|---|---|
| Score parts, out of 100 | ||
| Adoption | 96, widely used | 79, popular |
| Freshness | 100, active | 100, active |
| Maintenance | 91, 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 | 277.8k | 111.2k |
| License | MIT (permissive) | Apache-2.0 (permissive) |
| Last commit | Oct 2026 | Oct 2026 |
| Last release | Oct 2026 | Oct 2026 |
| Language | JavaScript | Go |
| Docker | Yes | Yes |
| GPU | Not needed | Not needed |
| arm64 or Apple Silicon | Not stated | Not stated |
ecc
ECC installs a set of 68 agents, 293 skills, 94 legacy command shims, hooks, rules, memory and continuous-learning tooling into coding agents, aiming to enforce a plan, test, implement, review, verify loop. It ships as a Claude Code plugin (`ecc@ecc`) and the `ecc-universal` npm package, with guided setup for Claude Code, Codex and Kimi Code. It also bundles AgentShield, which scans prompts, hooks, MCP config, permissions and secrets.
Who it is for: Developers using Claude Code or Codex who want preset workflows
Strengths
- Guided installer with dry-run for Claude Code, Codex and Kimi Code
- Bundles AgentShield scanner for hooks, MCP config, permissions and secrets
- Large catalog: 68 agents, 293 skills, plus selectable rule packs by language
- MIT license; release 2026-10-01 and commit 2026-10-02 are recent
Weaknesses
- Full support is Claude Code only; other harnesses are capability-limited adapters
- Stacking install methods duplicates skills and hooks; reset steps are needed
- Claude plugins cannot ship rules, so those are copied manually
- ECC Pro GitHub App for private repos is paid, from $19/seat/month
- no GPU
- Docker + Compose
- Needs Node.js 18+, Git, Claude Code 2.1+ (for plugin setup)
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.