Skip to content

Latest commit

Β 

History

History
112 lines (84 loc) Β· 3.64 KB

README.md

File metadata and controls

112 lines (84 loc) Β· 3.64 KB

πŸ›οΈ Delphi

Delphi is a versatile and adaptable conversational agent framework that streamlines chatbots and AI-driven agent systems. Its design prioritizes abstraction and provider-independence, allowing for seamless integration with various message providers. Delphi is compatible with a range of environments, including browsers, Node.js, and Deno, making it an ideal choice for diverse application scenarios.

Note: Delphi is currently in development. While usable, the API is subject to change. If you encounter a bug, please open an issue and help us improve the framework!

πŸš€ Features

  • πŸ€– Provider-Independent: Delphi abstracts the messaging function, allowing seamless integration with various providers like OpenAI and Azure OpenAI.
  • πŸ“ž Function Calling: The agent can call functions to perform tasks such as querying a database or making an API call, and incorporate the results into the conversation.
  • πŸ”„ Flexible Conversational Flow: Manages complex conversation patterns with ease, facilitating intelligent and context-aware interactions.
  • πŸ› οΈ Customizable Agent Options: Tailor your agent with configurable settings such as maximum rounds and specific model preferences.
  • 🌍 Cross-Platform Compatibility: Works effortlessly in browsers, Node.js, and Deno environments.
  • πŸ“ Schema Validation: Ensures the integrity and correctness of data with JSON schema validation.
  • πŸŽ›οΈ Modular Design: Facilitates easy integration and scalability through a clear and modular code structure.

πŸ“Œ Requirements

  • Node.js (v18 or newer), a modern browser, or Deno
  • TypeScript
  • An account with a supported message provider (e.g., Azure, OpenAI)

πŸ› οΈ Installation

Delphi is available on npm and can be easily installed:

npm install @wecandobetter/delphi

πŸš€ Getting Started

To quickly get started with Delphi:

  1. Import Delphi: Import the necessary components from Delphi in your project.

    import { Agent, Context } from "@wecandobetter/delphi";
  2. Create an Agent: Instantiate an Agent with your desired configuration.

    const agent = new Agent("myAgent" /* ... configuration ... */);
  3. Initialize Context: Set up a Context to manage the conversation state.

    const context = new Context();
    
    context.addMessage({
      role: "system",
      content: "You are a helpful assistant!",
    });
    
    context.addMessage({
      role: "user",
      content: "Hello, I need help.",
    });
  4. Run the Agent: Start the agent with the initialized context.

    for (const { message } of agent.run(context)) {
      console.log(message.content); // "Sure, how can I help?"
    }

This example gives you a basic setup for initiating a conversation with Delphi. For more detailed usage and advanced features, refer to the documentation.

πŸ“– Documentation

Visit the documentation for information on API usage and advanced features.

🀝 Contributing

Contributions are welcome! Create an issue, open a pull request, or reach out to us in the discussions section.

πŸ“œ License

Delphi is released under the MIT License. See the LICENSE file for more details.

βœ‰οΈ Contact

Reach out to us at contact@wcdb.life for support or inquiries.

Happy conversing! πŸŽ‰