Skip to content

πŸ”’ Safeguard LLM behavior with PromptGuard to detect unseen regressions and ensure reliable outputs amid evolving model updates.

Notifications You must be signed in to change notification settings

rizkycsv/PromptGuard

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

4 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

πŸš€ PromptGuard - Simplifying LLM Behavior Testing

πŸ“₯ Download PromptGuard

Download PromptGuard


πŸš€ Getting Started

PromptGuard is a tool that helps you ensure your language model delivers reliable results. By running the same prompts across different settings and comparing the outputs, you can catch issues before they affect users.

🎯 Features

  • Easy Comparison: Run tests on various model configurations.
  • Semantic Regression Alerts: Get notifications when model behavior changes unexpectedly.
  • User-Friendly Interface: Designed for easy navigation.
  • Open Source: Modify and contribute as needed.

πŸ’» System Requirements

To run PromptGuard, make sure your computer meets these minimum requirements:

  • Operating System: Windows, macOS, or Linux.
  • Python Version: 3.8 or higher.
  • Memory: At least 4 GB RAM.
  • Storage Space: Minimum of 100 MB available space.

πŸ“₯ Download & Install

  1. Visit the Releases page.
  2. Choose the latest version of PromptGuard listed.
  3. Select the appropriate file for your operating system.
  4. Download the file to your computer.
  5. Once the download is complete, locate the file in your downloads folder.
  6. Double-click the file to start the installation.
  7. Follow the on-screen instructions to complete the setup.

After installation, you can start using PromptGuard right away.

πŸ“– Usage Instructions

  1. Open PromptGuard: Find the application in your programs list and click to open.
  2. Set Up Your Test: Input your prompts and choose the model settings you want to test against.
  3. Run the Comparison: Click the β€œRun Test” button to start the process.
  4. Review Results: PromptGuard will display any differences and highlight issues so you can make necessary adjustments.

ℹ️ Support & Community

If you encounter any issues or have questions, check these resources:

  • Documentation: Access detailed guides for advanced features.
  • Community Forum: Join discussions with other users to share insights and tips.
  • GitHub Issues Page: Report bugs or request features directly.

πŸ› οΈ Contributing

If you would like to help improve PromptGuard, we welcome contributions! You can:

  • Submit a pull request for any feature or bug fix.
  • Report any bugs you find through the Issues page.
  • Suggest new features that would enhance the tool.

πŸŽ‰ Acknowledgments

Thanks to everyone who has contributed to this project. Your input helps us create a better tool for all users.

πŸ“œ License

PromptGuard is open source and available under the MIT License. Feel free to use it, modify it, and share it.

For more detailed information about usage, features, and more, refer back to the GitHub repository.

Releases

No releases published

Packages

No packages published

Contributors 2

  •  
  •  

Languages