[MODEL] Claude code keep ignoring the skills i provided , dowing stupid things, wating my tokens. when i asked it if he see the skill - it answers - yes. i ignored itWTF?

Preflight Checklist

• [x] I have searched existing issues for similar behavior reports
• [x] This report does NOT contain sensitive information (API keys, passwords, etc.)

Type of Behavior Issue

Claude modified files I didn't ask it to modify

What You Asked Claude to Do

Why claude keep ignoring my skills, make many mistakes, kill my tokens?
when i ask - why you did it this way and not used the skill - it replies - i did sloppy work.

I want my tokens back

What Claude Actually Did

ignored my skills over and over and over - i told it to keep it in MD files, i tried many different ways. half of the time it does stupid things

Expected Behavior

read the skills - work according to the skill

Files Affected

...

Permission Mode

Accept Edits was ON (auto-accepting changes)

Can You Reproduce This?

Yes, every time with the same prompt

Steps to Reproduce

_No response_

Claude Model

Sonnet

Relevant Conversation

Impact

Critical - Data loss or corrupted project

Claude Code Version

code

Platform

Anthropic API

Additional Context

_No response_