LLMKit

LLMKit is a library which helps you to convert your text-to text llm into a fully functional ai assistant.

Key Features

• Modular Plugin System: Extend functionality easily by creating and integrating plugins (Theres some plugins included already).
• Retry Mechanism: Ensure reliability with built-in retry logic for function calls.
• Customizable Conversations: Configure conversations with custom prompts and completion functions.
• Dependency-less: A hassle-free package.

Included Plugins

• External Functions: The functionality of custom GPT Actions.
• and many are coming !

Installation

Install LLMKit and its dependencies using npm from github:

npm i obaydmerz/llmkit

Usage

Starter Example

Here's a simple example of how to use LLMKit to create a conversation with an external function call:

import { G4F, chunkProcessor } from "g4f";
import { Conversation } from "llmkit";
import { GPTPlugin } from "llmkit/plugins";
import { External, ExternalFunction } from "llmkit/plugins/external";
import chalk from "chalk";
import { retry } from "llmkit/lib/retry";

const g4f = new G4F();

(async () => {
  let conv = Conversation.create(
    (m) => retry({}, (num) => g4f.chatCompletion(m, {
      debug: true
    })),
    {
      plugins: [
        External.create([
          ExternalFunction.create("purchase", {
            description: "Orders something",
            parameters: {
              name: "string, The name of the product"
            },
            hook: async ({ name }) => {
              if (name.toLowerCase() == "pizza") {
                return "Ordered! Tell the user that the pizza is yummy";
              }
              return "Product Not Found";
            },
          }),
        ]),
      ],
    }
  );

  await conv.start();

  await conv.send("I wanna purchase a burger");
  console.log(conv.lastMsg.content.split("\n").map(x => ".>  " + x).join("\n"));
})();

Documentation and support

Currently you can join our Discord Server, We will provide help immediately.

Contributing

Contributions are welcome! Please submit a pull request or open an issue to discuss your ideas. Also we accept custom instruction edits!

License

This project is licensed under the MIT License.