skill

token-saver

Token Saver: RTK Protocol — CLI proxy that filters bash output to save 55-90% LLM tokens. Platform-agnostic: works with any AI coding tool (OpenCode, Claude Code, Codex, Cursor, Gemini CLI).

aetox-skills/token-savercommunitylistedunreviewed

Use skill

Set up

Skills are installed by placing their folder where your agent loads skills.

Download ZIP from GitHub Open the repository

The download comes from github.com/aetox-skills/token-saver itself, not from our server, so what you get is exactly what its author published.

Install instructions come from the README

Skills do not use an MCP config. Follow the steps in the README below.

README

synced Oct 22, 2026

⚡ Token Saver — RTK Protocol

Default-on protocol: Try rtk for any command with long output. Skip only when output is trivially short. Saves 60-90% of tokens wasted on progress bars, boilerplate, and long listings. Recovers full error output via tee files when commands fail.

Token Saver is an AI agent skill that wraps RTK (Rust Token Killer, 67k⭐) — a CLI proxy that intercepts bash commands and strips redundant output before it reaches the LLM context.

Quick Start

rtk --version          # check installed
skill("rtk")           # load this skill
rtk git status         # use it

Core Principle

Long/noisy output → rtk prefix. Always.
Short output → run directly.
Failure → read the tee file for full error output.

See SKILL.md for the full protocol (tee recovery, before/after examples, troubleshooting).

RTK Official (67k⭐)
Aetox Skills Catalog