Comparison
Codex Memory Trim vs Caveman
Caveman has more GitHub stars (112,880 vs 101). Caveman was updated more recently (last commit 1d ago). Data comes straight from GitHub and refreshes daily.
Codex Memory Trim
给 Codex 全局记忆定期减负的 skill：检测重复、清理过期线程、压缩冗长表达、添加自定义全局规则 / A skill that puts Codex global memory on a periodic diet: dedupe, prune, compress, add global rules
$
git clone https://github.com/Yu-Xiao-Sheng/codex-memory-trim
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.
$
/plugin marketplace add JuliusBrussee/caveman| Metric | Codex Memory Trim | Caveman |
|---|---|---|
| Type | Agent skill | Agent skill |
| GitHub stars | 101 | 112,880 |
| Stars, last 7 days | n/a | +481 |
| Forks | 1 | 6,531 |
| Upvotes | 0 | 0 |
| Language | Shell | Go |
| License | MIT | Apache-2.0 |
| Last commit | 1mo ago | 1d ago |
| Created | Aug 2026 | Apr 2026 |
| Categories | AI & LLM tooling, Memory & knowledge | AI & LLM tooling |