Token Efficiency

This is the skill you add when you're tired of burning tokens on log files and unnecessary file reads. It teaches Claude to use grep instead of reading entire files, bash commands instead of the read-edit-write cycle, and quiet flags by default. The model selection guidance alone (Opus for learning codebases, Sonnet for everything else) can cut costs in half. The real value is in the decision trees: when to use sed versus Read+Edit, when bash makes sense versus when you actually want to see a diff. It's opinionated about progressive disclosure and includes a helpful myth-buster about skill loading costs. If you've ever watched Claude read a 10,000 line log file to find one error, you need this.

skills.sh installs: 2.6K
GitHub stars: 23
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add delphine-l/claude_global --skill token-efficiency --agent claude-code

Installs into .claude/skills of the current project.

Files

Token Efficiency Expert

This skill provides token optimization strategies for cost-effective Claude Code usage across all projects. These guidelines help minimize token consumption while maintaining high-quality assistance.

Core Principle

ALWAYS follow these optimization guidelines by default unless the user explicitly requests verbose output or full file contents.

Default assumption: Users prefer efficient, cost-effective assistance.

Model Selection Strategy

Use the right model for the task to optimize cost and performance:

Opus - For Learning and Deep Understanding

Use Opus when:

Learning new codebases - Understanding architecture, code structure, design patterns
Broad exploration - Identifying key files, understanding repository organization
Deep analysis - Analyzing complex algorithms, performance optimization
Reading and understanding - When you need to comprehend existing code before making changes
Very complex debugging - Only when Sonnet can't solve it or issue is architectural