prompt__reverse-engineering

Extract underlying instructions from LLM output into reusable system prompts.

1|1|Updated Mar 16, 2026

Search Tags:#template #analysis #LLM #quality #prompt-engineering #reverse-engineering #system-prompt

Ready to run

npx skills add https://github.com/erikhazzard/vasir --skill prompt-reverse-engineering

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: prompt__reverse-engineering
Source: https://github.com/erikhazzard/vasir/tree/main/.agents/skills/prompt__reverse-engineering
Command: npx skills add https://github.com/erikhazzard/vasir --skill prompt-reverse-engineering

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured approach to reverse-engineering a system prompt from high-quality LLM output. It captures the decisions, constraints, and conventions that shaped the result so they can be codified into a reusable template.

Core Features & Use Cases

Extract the underlying instruction patterns, constraints, and role definitions from a successful output.
Synthesize a canonical system prompt that generalizes to new inputs while preserving quality and style.
Create reusable prompt templates for analytic, code-review, or documentation tasks across different domains.
Use Case: When a high-quality response is achieved, generate a generalized system prompt that reproduces the method on new data.

Quick Start

Provide a fresh deliverable and instruct the system to generate a paste-ready SYSTEM PROMPT that reproduces the same category of work for entirely new inputs.

Frequently Asked Questions about prompt__reverse-engineering

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

FAQPage Schema

How do I reverse-engineer a system prompt from high-quality LLM output?▼

Reverse-engineering a prompt from LLM output extracts the underlying instructions, constraints, and role definitions that shaped the result. It synthesizes a canonical, generalizable system prompt suitable for reproducing that quality across new contexts.

What is the best way to create reusable prompt templates from previous LLM responses?▼

Creating reusable prompt templates requires analyzing high-quality LLM outputs to codify their underlying decisions and conventions. The synthesized template enforces methodical analysis and evidence-based reasoning while preserving the original style for new data.

Can I extract a prompt template for code review tasks from an existing output?▼

Yes, you can extract a prompt template for code review tasks from existing outputs. The reverse-engineering process applies to analytical, coding, or documentation tasks, capturing constraints to create templates for new inputs.

How does reverse-engineering system prompts preserve output style and constraints?▼

Reverse-engineering system prompts preserves style by extracting the underlying instruction patterns from a successful deliverable. It ensures the reconstructed prompt enforces explicit constraints and methodical analysis for new contexts.

Do I need a specific framework to reverse-engineer prompts for analytical tasks?▼

No specific framework is needed to reverse-engineer prompts for analytical tasks. You provide a fresh deliverable, and the system generates a self-contained, generalizable system prompt without external dependencies.