Skip to content

Latest commit

 

History

History
153 lines (117 loc) · 4.9 KB

README.md

File metadata and controls

153 lines (117 loc) · 4.9 KB
Raw

Baymax

Baymax is a Slack bot powered by OpenAI's GPT models, designed to assist users within Slack channels. It listens to messages where it's mentioned, processes them using OpenAI's Chat Completion API, and responds accordingly. Baymax supports plugin extensions, allowing for custom tool integrations.

Features

  • Slack Integration: Listens and responds to messages in Slack channels.
  • OpenAI GPT Integration: Uses OpenAI's GPT models for generating responses.
  • Plugin Support: Load custom plugins to extend functionality.
  • NATS Messaging: Utilizes NATS for internal messaging between services.
  • Health Checks: Provides liveness and readiness endpoints.
  • Docker and Helm Support: Easily deployable using Docker and Kubernetes Helm charts.

Table of Contents

Architecture Overview

Baymax consists of the following components:

  • cmd/baymax: The main application entry point.
  • pkg/providers: Providers for Slack, OpenAI, and NATS services.
  • pkg/services: Business logic for handling Slack and OpenAI interactions.
  • pkg/transport: Event handlers and transport mechanisms.
  • pkg/plugins: Plugin system for extending Baymax functionalities.
  • chart/baymax: Helm chart for deploying Baymax on Kubernetes.

Installation

Prerequisites

  • Go: Version 1.22 or later.
  • Docker: For containerization (optional).
  • Kubernetes: Cluster for deployment (optional).
  • Slack App: With proper tokens.
  • OpenAI API Key: For accessing OpenAI services.
  • NATS Server: Optional, Baymax can start its own embedded server.

Clone the Repository

git clone https://github.com/fandujar/baymax.git
cd baymax

Build from Source

go build -o baymax ./cmd/baymax

Using Docker

Build the Docker image:

docker build -t baymax:latest -f docker/Dockerfile .

Configuration

Baymax requires the following environment variables:

  • SLACK_APP_TOKEN: Slack App-level token (starts with xapp-).
  • SLACK_BOT_TOKEN: Slack Bot token (starts with xoxb-).
  • OPENAI_API_KEY: OpenAI API key.

Optional environment variables:

  • LOG_LEVEL: Defines de log level for the application.
  • BAYMAX_NAME: The name Baymax will introduce itself as (default: "Baymax").
  • BAYMAX_PLUGINS_DIR: Directory to load plugins from (default: current directory).
  • OPENAI_SYSTEM_MESSAGE: Defines a custom system role message.
  • OPENAI_MODEL: Defines the OpenAI Model to be used. Defaults to gpt-4o-mini.

Set these variables in your environment or pass them when running the application.

Usage

Running Locally

Set the required environment variables and run:

./baymax

Running with Docker

docker run -e SLACK_APP_TOKEN=your_app_token \
           -e SLACK_BOT_TOKEN=your_bot_token \
           -e OPENAI_API_KEY=your_openai_key \
           baymax:latest

Deploying to Kubernetes with Helm

First, create a Kubernetes secret with your tokens:

kubectl create secret generic baymax-secrets \
  --from-literal=slack-app-token=your_app_token \
  --from-literal=slack-bot-token=your_bot_token \
  --from-literal=openai-api-key=your_openai_key

Install the Helm chart:

helm install baymax ./chart/baymax

Plugin Development

Baymax supports plugins to extend its functionality. Plugins should implement the Plugin interface defined in pkg/plugins/plugins.go. Compile your plugin as a Go plugin (.so file) and place it in the directory specified by BAYMAX_PLUGINS_DIR.

Plugin Interface

type Plugin interface {
    GetTools() []openai.Tool
    RunTool(toolName string, parameters string, messages []openai.ChatCompletionMessage, tools []openai.Tool) (string, error)
    RunEventLoop(natsClient *nats.Conn)
}

Health Checks

Baymax exposes health check endpoints:

  • Liveness Probe: /liveness on port 8081.
  • Readiness Probe: /readiness on port 8081.

Logging

Baymax uses zerolog for structured logging. The log level is set to Info by default.

Contributing

Contributions are welcome! Please open an issue or submit a pull request on GitHub.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Contact

For questions or support, please open an issue on the GitHub repository.