Documentation
Quick Start Guide
Article updated: March 20, 2023 · Reading time: 5 min
This quick start guide will help you set up your PromptScout account, create your first prompt, and start building your prompt library in just a few minutes.
Prerequisites
Before you begin, make sure you have:
- Created a PromptScout account (see Creating Your Account)
- A basic understanding of what you want to accomplish with your prompts
- Optional: API keys for your preferred LLM providers if you want to test prompts directly in PromptScout
5-Minute Setup Guide
Sign in to your account
Navigate to promptscout.com/login and sign in using your email/password or Google account.
Complete your profile
Click on your profile icon in the top right corner and select "Profile Settings". Add your display name and select your default preferences for:
- Default LLM (e.g., Claude, GPT-4, etc.)
- Theme preference (light/dark/system)
- Dashboard layout (grid/list view)
Optional: Configure API keys
If you want to test prompts within PromptScout, go to Settings > API Keys and add your API keys for the LLM providers you use. PromptScout securely encrypts all API keys.
Create your first prompt
Click the "+ New Prompt" button in the top navigation bar or dashboard. In the prompt editor:
- Give your prompt a descriptive name
- Choose the prompt type (user or system)
- Select your target LLM
- Write your prompt content or select a template
- Add tags for organization (optional)
- Click "Save" to add it to your library
Test your prompt
Use the testing panel to try out your prompt:
- Enter a sample input in the test field
- Click "Test" to see the response (requires API key configuration)
- Review the output and refine your prompt as needed
- Save any changes to update your prompt
Next Steps: Building Your Prompt Library
Now that you've created your first prompt, here are some suggestions for building out your library:
Create category-specific prompts
Create prompts for different purposes such as content writing, data analysis, coding assistance, or creative writing. Use tags to organize them by category.
Experiment with different LLMs
Try creating the same prompt for different LLMs and compare the results. Different models have different strengths and may perform better for specific tasks.
Explore template variations
Use PromptScout's built-in templates as starting points and modify them to suit your specific needs. Save variations of successful prompts to refine your approach.
Establish a tagging system
Create a consistent tagging system to organize your prompts. Consider tags for purpose, domain, complexity, or target audience to make finding the right prompt easier.
Sample Prompt Templates
Here are a few simple prompt templates to get you started:
Content Summarizer (User Prompt)
{text_to_summarize}
Provide the summary in bullet points and include a one-sentence TL;DR at the end.
Research Assistant (System Prompt)
1. Provide factual information backed by reputable sources
2. When citing research, include author, year, and key findings
3. Acknowledge areas of uncertainty or debate in the field
4. Avoid jargon and explain complex concepts clearly
5. Offer balanced perspectives on controversial topics
If you don't know the answer to a question, say so rather than speculating.
Code Reviewer (User Prompt)
1. Bugs and potential issues
2. Performance optimizations
3. Readability improvements
4. Security vulnerabilities
5. Adherence to best practices
{code_to_review}
Provide specific suggestions for improvements with explanations for each recommendation.
Ready to go deeper?
Now that you have the basics, explore these resources to enhance your prompt engineering skills:
- System vs User Prompts - Learn when and how to use each prompt type
- Advanced Prompt Techniques - Discover strategies for more effective prompts
- Analyzing Results - How to interpret and improve prompt performance
Related Articles
Creating Your First Prompt
Detailed steps for crafting an effective prompt with PromptScout.
Read articleSystem vs User Prompts
Understand the differences between system and user prompts and when to use each.
Read article