Prompt Engineer

Download SKILL.mdGet Murage How skills work

Open source. macOS, Windows and Linux.

At a glance

What it is

Prompt engineering mastery covering chain-of-thought reasoning, few-shot learning, system prompt design, structured output with JSON mode, prompt injection prevention, temperature and top-p tuning, context window management, evaluation metrics, and A/B testing prompts.

A skill is a written procedure a bot loads when a job calls for it. This one is a single file, SKILL.md, and the whole file is on this page.

File: SKILL.md
Length: 537 lines2,117 words, about 10 min read
Category: AI and machine learningLLM engineering
Level: Intermediate
License

More AI and machine learning skills.

The full skill.

The whole file, exactly as it downloads. Read it before you rely on it.

SKILL.md

537 lines · 2,117 words

Download

Prompt Engineer

Overview

Prompt engineering is the discipline of designing, testing, and optimizing inputs to large language models to produce reliable, high-quality outputs. This skill covers the full spectrum from basic prompt construction to advanced techniques including chain-of-thought reasoning, structured output enforcement, injection prevention, and systematic evaluation.

Core Prompting Techniques

Zero-Shot Prompting

Direct instruction without examples. Effective for well-understood tasks.

Put this skill to work.

Questions.

What is the Prompt Engineer skill?

How much does it cost?

Nothing. It ships with Murage, which is open source. Your bots run on the AI plan you already have.

Can I change it?

Yes. Open it in Settings → Skills to read, edit or duplicate it. Skill Guard scans it again after every edit.

Give your first job to Murage.

Download the app, connect the AI you already pay for, and tell your Chief of Staff what needs doing. Plan on about ten minutes from install to a working team.

Download Murage

No account needed · Runs on the AI plan you already pay for

Variant	Description	Best For
Standard CoT	"Let's think step by step"	Math, logic
Zero-shot CoT	Appending reasoning trigger	General reasoning
Self-consistency	Multiple CoT paths, majority vote	High-stakes decisions
Tree of Thought	Branching exploration	Complex planning

Parameter	Range	Low Value Effect	High Value Effect
Temperature	0.0 - 2.0	Deterministic, focused	Creative, diverse
Top-p	0.0 - 1.0	Restricts token pool	Wider token pool
Top-k	1 - N	Very focused	More variety
Frequency penalty	-2.0 - 2.0	Allows repetition	Penalizes repetition
Presence penalty	-2.0 - 2.0	Allows topic repetition	Encourages new topics

Anti-Pattern	Problem	Fix
Vague instructions	Inconsistent outputs	Be specific about format, length, style
No output format	Unparseable results	Specify exact format with examples
Information overload	Key details buried	Prioritize, use headers, keep focused
Negative instructions only	Model unsure what TO do	State positive instructions first
No error handling	Silent failures	Define fallback behavior explicitly
Prompt stuffing	Token waste	Trim to essential context only

Prompt Engineer

More AI and machine learning skills.

The full skill.

Prompt Engineer

Overview

Core Prompting Techniques

Zero-Shot Prompting

Put this skill to work.

Questions.

Give your first job to Murage.

Search murage.ai

Prompt Engineer

More AI and machine learning skills.

The full skill.

Prompt Engineer

Overview

Core Prompting Techniques

Zero-Shot Prompting

Put this skill to work.

Questions.

Give your first job to Murage.

Create a bot and use this skill.

Ask Ember to create the bot

Open the bot’s skills

Find Prompt Engineer and switch it on

Ask the bot for the job

About this skill.

Few-Shot Prompting

Chain-of-Thought (CoT)

Self-Consistency Prompting

System Prompt Design

Anatomy of an Effective System Prompt

Production System Prompt Example

Structured Output (JSON Mode)

Enforcing JSON Output

JSON Schema Enforcement

Prompt Injection Prevention

Threat Categories

Defense Strategies

Input Sanitization

Delimiter-Based Isolation

Temperature and Sampling Parameters

Parameter Reference

Decision Matrix

Context Window Management

Token Budget Planning

Strategies for Long Inputs

Evaluation Metrics

Automated Evaluation

LLM-as-Judge Evaluation

A/B Testing Prompts

Framework

Sample Size Calculation

Advanced Techniques

Prompt Chaining

Meta-Prompting

Common Anti-Patterns

Checklist

When to Use

Example

Edge Cases