correct-skill

Encode operator corrections as durable Hard Rules in SKILL.md files.

1|Updated Apr 20, 2026
One-click install
npx skills add https://github.com/Largo2z9/phantomos --skill correct-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: correct-skill
Source: https://github.com/Largo2z9/phantomos/tree/main/.skills/skills/correct-skill
Command: npx skills add https://github.com/Largo2z9/phantomos --skill correct-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

## What problem does it solve? This Skill prevents repeated mistakes by encoding operator corrections directly into existing skill instructions as durable Hard Rules.

## Core Features & Use Cases

  • Correction Encoding: Converts identified skill failures into permanent rules inside the target SKILL.md contract.
  • Safe Skill Patching: Reviews existing instructions, validates rule placement, and preserves the skill structure during updates.
  • Use Case: When a skill generates an incorrect output repeatedly, use this Skill to capture the exact correction so future invocations avoid the same failure.

### Quick Start Ask the correct-skill skill to patch a named skill with a hard rule explaining the observed error and required correction.

Frequently Asked Questions about correct-skill

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

FAQPage Schema
How do I permanently encode AI agent corrections into workspace skill files?

To encode AI agent corrections into workspace skill files, you apply operator feedback directly into the SKILL.md contract as durable Hard Rules to prevent repeated incorrect behavior.

What is instruction patching for AI agents and when should I use it?

Instruction patching for AI agents is the process of converting identified skill failures into permanent rules inside target SKILL.md files, used when a skill repeatedly generates incorrect outputs.

How do I patch a SKILL.md file to add hard rules without breaking the existing structure?

To safely patch a SKILL.md file, the skill reviews existing instructions, validates rule placement, and preserves the skill structure during updates to ensure the contract remains intact.

Do I need to validate SKILL.md structure before modifying skill contracts?

Yes, you must validate the SKILL.md structure before modifying skill contracts to ensure safe skill patching and preserve the existing workspace skill file definitions during the update.

Why does my AI agent keep making the same mistake despite prompt engineering adjustments?

AI agents keep making the same mistake because temporary prompt engineering adjustments are not durable, requiring the encoding of operator corrections as hard rules into the skill definition.

Can I automate workspace skill maintenance to prevent repeated incorrect behavior?

Yes, you can automate workspace skill maintenance by capturing exact corrections and encoding them into existing skill instructions, ensuring future invocations avoid the same failure.