BabyAGI JS is a JavaScript-based AI agent that creates, prioritizes, and executes tasks using the GPT 3.5 or GPT 4 architecture. It integrates with OpenAI's language model to create a powerful AI that can handle a wide range of tasks.
- Task creation: Generates new tasks based on the current context and objectives.
- Task prioritization: Reorders tasks according to their importance and relevance to the main objective.
- Task execution: Performs tasks and returns results.
- Clone this repository.
- Add API keys to your .env
OPENAI_API_KEY=
- Install the required dependencies using
npm install
. - Write your code in the
src
directory. - Run your program with
npm run start
.
This file initializes the BabyAGI agent with the required configurations, including the language model and objective. It imports the BabyAGI
class from babyagi.js
and creates a new instance to perform tasks based on the given objective.
This file contains the core implementation of the BabyAGI agent. It defines three main classes, TaskCreationChain
, TaskPrioritizationChain
, and ExecutionChain
, which are responsible for creating, prioritizing, and executing tasks, respectively.
The BabyAGI
class combines these three classes and provides methods to add tasks, print tasks, and execute tasks. The call
method is the main entry point to start the agent's task processing loop.
The following is an example of how to use the BabyAGI agent:
- Set the objective in
src/index.ts
:
const OBJECTIVE = 'Integrate stripe in typescript';
- Run the program with
npm run start
. - The BabyAGI agent will create, prioritize, and execute tasks based on the given objective.
We welcome contributions to improve BabyAGI JS. Feel free to open an issue or submit a pull request.
This project is licensed under the MIT License.