Writeup/Architecture/Original by Mehansh

Skills, Not Prompts

The essential difference lies in triggers and negative constraints.

What It Does

Explores why traditional prompts fail in production workflows, and how explicit trigger declarations, negative constraints, and output rubrics transform prompts into reliable skills.

Architecture and Design Rationale

Prompts provide general guidance without boundaries. Skills define when to activate, what to refuse, and what a passing output requires.

When to Activate

•Authoring new agent skills or evaluating model reliability

Skill Specification and Runtime Setup

Full Instruction Specification

# Skills, Not Prompts Specification

## Overview
Explores why traditional prompts fail in production workflows, and how explicit trigger declarations, negative constraints, and output rubrics transform prompts into reliable skills.

## Rationale
Prompts provide general guidance without boundaries. Skills define when to activate, what to refuse, and what a passing output requires.

## Core Principles
* Always verify environment constraints and dependencies before executing.
* Enforce negative constraints and reject speculative abstractions.
* Validate output against stated acceptance rubrics.

## Execution Workflow
1. Discovery: Map requirements and inspect active context.
2. Decomposition: Break down the task into discrete, verifiable phases.
3. Execution: Implement surgical changes preserving existing code architecture.
4. Verification: Test all criteria before declaring completion.