Treat

Skill Verified Active

Prune bloated session with a prescription. Removes progress ticks, stale reads, duplicate content, and more.

Purpose

To significantly reduce the token count and size of LLM sessions, making API calls more efficient and cost-effective, while preserving essential context and conversation integrity.

Features

18 composable pruning strategies
Continuous session protection via guard daemon
Behavioral digest for persisting corrections
Agent Teams state protection
Token-aware diagnostics and context window auto-detection
Multiple installation methods (pip, pipx, Homebrew, Nix)

Use Cases

Reducing API costs by minimizing token usage in long LLM sessions
Preventing context loss and coordination issues in Agent Teams
Diagnosing and fixing session corruption or performance issues
Maintaining lean and efficient LLM sessions for developers

Non-Goals

Performing actions not directly related to session content pruning
Modifying conversation history or agent behavior beyond extracting corrections
Replacing the core LLM functionality or session persistence mechanisms
Operating without explicit user confirmation for destructive actions

Installation

First, add the marketplace

/plugin marketplace add Ruya-AI/cozempic

/plugin install plugin@cozempic

Quality Score

Verified

100 /100

Analyzed about 24 hours ago

Trust Signals

Last commit1 day ago

GitHub owner Ruya-AI

Stars295

LicenseMIT

Websitepypi.org

Status

View Source

Similar Extensions

Guard

100

Protect Claude Code sessions from context overflow by running a background daemon that monitors session size and auto-prunes before compaction hits. Use when the user says "guard", "protect session", "context getting long", "prevent compaction", "session management", or is running agent teams that need continuous context protection.

Skill

Ruya-AI