skills.rest
hiddink-ai
hiddink-aiOFFICIAL
Explore all skills by this author →
GitHub

token-efficiency-audit

Audit and optimize AI agent settings to minimize token consumption.

1|Updated May 21, 2026
Search Tags:#optimization#configuration#efficiency#claude-code#token#cost-management
One-click install
npx skills add https://github.com/hiddink-ai/hiddink-harness --skill token-efficiency-audit
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: token-efficiency-audit
Source: https://github.com/hiddink-ai/hiddink-harness/tree/main/templates/skills/token-efficiency-audit
Command: npx skills add https://github.com/hiddink-ai/hiddink-harness --skill token-efficiency-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This skill addresses the challenge of excessive token consumption in AI coding sessions by identifying and disabling unnecessary context injections and configuring optimal output limits.

Core Features & Use Cases

  • Three-Layer Defense: Coordinates cache protection, runtime compression, and pre-session configuration to minimize overhead.
  • Interactive & CI Modes: Provides safe, interactive configuration for developers and strict, automated settings for CI/worker pipelines.
  • Status Monitoring: Offers a comprehensive audit of current settings against recommended efficiency floors to prevent re-call loops.

Quick Start

Run the token-efficiency-audit skill to audit your current project settings and identify potential token savings.

Frequently Asked Questions about token-efficiency-audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I reduce token consumption in Claude Code?▼

To reduce token consumption in Claude Code, run a token-efficiency audit to identify and disable unnecessary context injections, applying optimal output limits and efficiency guardrails to your project configuration settings.

What is the best way to prevent AI context window overflow during coding sessions?▼

Preventing AI context window overflow is achieved by applying a three-layer defense: cache protection, runtime compression, and pre-session configuration to minimize overhead and enforce efficiency limits.

How do I optimize AI agent configuration settings for cost management?▼

You optimize AI agent configuration settings by auditing current project environments against recommended efficiency floors, modifying .claude/settings.json and .claude/settings.local.json to apply performance levers.

Can I apply token efficiency settings automatically in CI pipelines?▼

Yes, you can apply token efficiency settings automatically in CI pipelines using strict, automated configuration modes designed for worker pipelines, alongside interactive modes for local developer environments.

Does the token optimization audit work with Codex CLI?▼

Yes, the token optimization audit works with Codex CLI by applying optimal output limits and efficiency guardrails to both project-level and local environment settings to minimize overhead.

Related Skills

Explore More Software Engineering Skills→
hendrax5hendrax5

claude-code-learning

Analyze CLAUDE.md and .claude configurations to generate recommended settings.

Community
Intermediate
mintuzmintuz
21

prompt-master

Transform simple prompts into XML-tagged instruction sets with roles, steps, and examples.

Community
Basic
viktorbezdekviktorbezdek
10

context-compression

Compress LLM conversation context using anchored iterative summarization.

Community
Advanced
goodnessibehgoodnessibeh
1

context-compression

Compress AI agent context to stay within token limits while preserving key information.

Community
Advanced