Treat

When your Claude session gets bloated with progress ticks, duplicate file reads, and old tool output, this prunes it down without losing conversation continuity. You get three prescriptions: gentle (40-55% savings), standard (50-70%), or aggressive (70-95%), each with different strategies like collapsing thinking blocks or trimming mega-blocks. It always dry-runs first so you see exactly what gets cut and the token savings before anything changes. The main thing to know is you probably want the reload skill instead, since it does the same prune but auto-resumes in a fresh terminal. Use treat only if you want to stay in your current session or handle the resume yourself.

GitHub stars: 332
last updated: Jun 11, 2026

View on GitHub

Install

$ npx -y skills add ruya-ai/cozempic --skill treat --agent claude-code

Installs into .claude/skills of the current project.

Files

Apply a pruning prescription to the current session. Default is standard if no argument given.

When to use treat vs reload: prefer the reload skill — it does the same prune plus auto-resume in one step (/exit → new terminal opens with the pruned session). Use treat only when the user explicitly wants to stay in the current session, work in a multi-pane setup, or resume manually.

Steps

Diagnose first — show the user what they're working with:
```
cozempic current --diagnose
```
Dry-run the treatment — show savings without applying:
```
cozempic treat current -rx $ARGUMENTS
```
If no argument was provided, use standard:
```
cozempic treat current -rx standard
```
Show results — present the dry-run output including token savings (the Tokens: line). Always surface both byte and token savings.
Ask confirmation — use AskUserQuestion to confirm before applying.

Apply on confirmation:

cozempic treat current -rx $ARGUMENTS --execute

SKILL.md on GitHub

Details

Repository: github.com/ruya-ai/cozempic
License: MIT
First seen: Jun 11, 2026

More from ruya-ai/cozempic

All 5