opt

Optimize prompts and system messages through a multi-phase PromptForge pipeline.

1|Updated Apr 12, 2026
One-click install
npx skills add https://github.com/cryptopafi/nexusos-skills --skill opt
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opt
Source: https://github.com/cryptopafi/nexusos-skills/tree/main/opt
Command: npx skills add https://github.com/cryptopafi/nexusos-skills --skill opt

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill converts rough or unclear prompts, system messages, and instructions into clearer, more complete, and agent-appropriate prompts to reduce wasted iterations and improve model output quality.

Core Features & Use Cases

  • Classification Gate: Automatically classifies input as TRIVIAL, STANDARD, COMPLEX, or PRODUCTION to select the appropriate optimization path.
  • Technique Selection & Multi-Phase Pipeline: Chooses prompt-engineering techniques and runs a multi-phase PromptForge pipeline (SCOPE, Optimize, Structure, Score, Meta-Prompting) to produce a polished prompt and a 0–100 score breakdown.
  • Cortex Integration & Persistence: Pre-searches and optionally persists high-scoring variants to a Cortex collection for reuse (PRODUCTION flow); includes explicit error handling and fallbacks for unavailable resources.
  • Use Case: Improve a customer-service system prompt, refine a technical code-generation instruction, or optimize an agent brief for production deployment.

Quick Start

Use the /opt command to optimize the prompt "Create a concise and persuasive product description for a new wireless earbud".

Frequently Asked Questions about opt

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

FAQPage Schema
How do I optimize a system prompt for better AI output quality?

Optimizing a system prompt involves classifying input complexity and applying a multi-phase pipeline that refines clarity, completeness, and agent suitability to achieve a scored, high-impact instruction.

What is the best way to refine raw prompts for production deployment?

The best way to refine raw prompts is using a PRODUCTION classification flow that applies multi-phase optimization, scores the result 0-100, and optionally persists high-scoring variants to a Cortex collection for reuse.

How does prompt classification improve prompt engineering results?

Prompt classification improves engineering results by automatically categorizing inputs as TRIVIAL, STANDARD, COMPLEX, or PRODUCTION, which selects the appropriate optimization path and specific techniques for each scenario.

Can I use this to optimize technical code-generation instructions?

Yes, you can optimize technical code-generation instructions. The Skill handles casual, creative, technical, and production scenarios, refining agent briefs and instructions to reduce wasted iterations and improve model output.

Why does my prompt score low on clarity and completeness?

Your prompt scores low because the optimization pipeline evaluates clarity, completeness, correctness, and focus using a 0-100 scoring breakdown, identifying structural weaknesses in raw prompts, system messages, or agent briefs.

When should I not use an automated prompt optimization pipeline?

You should avoid automated prompt optimization pipelines for trivial requests that require no structural refinement, as the classification gate will route them differently to avoid unnecessary processing overhead.