A simple open-source prompt management platform.

Explore the docs »

Available in: Python

⭐️ Why PromptDesk?

PromptDesk is 100% free and open-source minimalist tool to facilitate the creation, organization, integration, and evaluation of prompts, prompt-based applications, agents and Large Language Models (LLMs).

We recommend starting with the Quickstart guide. You can also jump straight to the Build Prompts or start integrating prompts in python with Integrate Prompts guides.

✨ Features

• 👨‍💻 OpenAI-like minimalist prompt builder
• 🥸 Prompt variable and logic support with Handlebars.js syntax
• 📊 Complete audit log of all API calls
• 🌐 Unlimited vendor-agnostic LLM API integrations
• 📦 Environment variables support
• 🚀 Easy to set up and integrate (5 minutes)
• 👨‍💻 Community-driven

🚀 Getting Started

PromptDesk is available as a hosted web application and as a Docker image. You can get started immediately by visiting the PromptDesk Login and creating an account. Hosting is provided for free by the MongoDB and DigitalOcean Startup Programs.

Hosted Web Application

wget https://raw.githubusercontent.com/promptdesk/promptdesk/main/quickstart/docker-compose.yml

Start the Docker container:

docker compose up

Open your web browser and navigate to http://localhost:8080/ to access the PromptDesk application. Initial setup will require a valid OpenAI API key for installation testing purposes.

👨‍💻 Python Integration

To install PromptDesk, use pip:

pip install promptdesk

Usage

You can call the prompt you build in the application by using the generate method.

from promptdesk import PromptDesk

pd = PromptDesk(
    #defaults to os.environ.get("PROMPTDESK_API_KEY")
    api_key = "YOUR_PROMPTDESK_API_KEY"
    #defaults to https://app.promptdesk.ai
    service_url = "http://localhost:8080"
)

story = pd.generate("short-story", {
    "setting": "dark and stormy night",
    "character": "lonely farmer",
    "plot": "visited by a stranger"
})

print(story)

🔗 Important Links

For more information about PromptDesk, please refer to the following resources:

• Documentation
• Application GitHub Repository
• Python GitHub Repository
• PyPI Package
• Docker Hub
• Official Website