clean-skill

Audit Claude Code skill session logs to identify inefficient file-read patterns.

30|12|Updated Jun 21, 2026

Search Tags:#automation #performance #context #optimization #audit #claude-code

Ready to run

npx skills add https://github.com/anthony-chaudhary/fak --skill clean-skill-anthony-chaudhary

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: clean-skill
Source: https://github.com/anthony-chaudhary/fak/tree/main/.claude/skills/clean-skill
Command: npx skills add https://github.com/anthony-chaudhary/fak --skill clean-skill-anthony-chaudhary

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill addresses the issue of bloated, slow, or expensive AI agent sessions caused by inefficient context management in custom skills.

Core Features & Use Cases

Context Auditing: Analyzes session logs to identify which tool results consume the most context budget.
Optimization Proposals: Traces large reads back to specific instructions and proposes Python-based bundling helpers to replace bulk file reads.
Use Case: If a skill frequently reads multiple large markdown files, this tool audits the usage and suggests a helper script to project only the necessary data, significantly reducing token consumption.

Quick Start

Run the clean-skill command followed by the name of the skill you wish to audit to generate a context efficiency report.

Frequently Asked Questions about clean-skill

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

FAQPage Schema

How do I audit Claude Code skill context usage to reduce token consumption?▼

To audit Claude Code skill context usage, this skill analyzes session logs to identify inefficient file-read patterns and tool results that consume the most context budget. It traces large reads back to specific instructions to pinpoint token waste.

What is the best way to optimize AI agent sessions that are bloated from reading large markdown files?▼

The best way to optimize bloated AI agent sessions is to use Python-based bundling helpers that project only necessary JSON data. This skill proposes these helpers to replace bulk file reads and significantly reduce token consumption.

Do I need Python 3 to analyze session logs for context optimization?▼

Yes, you need Python 3 installed to perform context optimization. It is required for parsing session logs and performing JSON-based data projection to minimize token waste during the audit process.

How does context auditing identify which tool results consume the most context budget?▼

Context auditing identifies budget-heavy tool results by analyzing session logs to trace large file reads back to specific skill instructions. This mechanism highlights inefficient patterns and flags the exact tool outputs causing token bloat.

Can I use this context audit on both project-local and user-global automation skills?▼

Yes, you can apply this context audit to both project-local and user-global automation skills. The tool is designed for the maintenance and performance tuning of custom skills across different scopes.