Context Engineering Prompt Enhancer

Presets:

Load a saved configuration

Essential

1. Core Task / Base Prompt *

What is the essential job you need the AI to perform?

2. Context / Background

Fill in Core Task and Context to enable Auto-Enhance.

3. Role/Persona Assignment

Tells the AI *who* to be to focus its knowledge.

4. Target Audience

Who is this output intended for?

Style & Constraints

5. Output Style & Constraints

Tone/Clarity: Motivational tone, 8th-grade language, actionable steps. Format: Output as JSON array of objects. Concise/Direct: Brief, focused responses without filler or repetition. Include Examples: Provide concrete examples to illustrate points.

Add Custom Constraint:

6. Negative Constraints (Do NOT)

No Jargon: Avoid buzzwords and overly technical language. No Hallucination: Don't invent facts or statistics. No Passive Voice: Use active voice only. No Hype: Give critical realistic feedback, not excessive enthusiasm.

Add Custom "Do NOT":

7. Think Step-by-Step

Plan and Refine: Outline arguments, review constraints, then generate. Step-by-Step: Think through steps internally (hide work).

Custom Reasoning:

8. Ask Before Assuming

Ask for Missing Info: Request data instead of inventing it. Clarify Ambiguity: Ask before assuming interpretation.

Custom Checks:

Advanced Options

(click to expand)

These options fine-tune AI behavior. Most prompts work well without them.

9. Show Examples

Provide examples (good/bad) to teach the AI what to imitate or avoid.

10. Output Format

Provide a skeleton structure for the AI to follow.

11. Success Criteria

What makes a "good" response? Define success metrics.

12. When Rules Conflict

Which constraints matter most when there's a conflict?

Enhanced Prompt Output

0 characters | ~0 tokens (estimated)

Token estimate: chars ÷ 4

Target Model:

Test in:

AI Critique

Sign Up

Password

Already have an account? Sign in

Settings

OpenAI API Key

Your API key is stored locally in your browser and never sent to any server except OpenAI.