Herobot is your 24/7 customer service assistant that helps you manage multi-channel customer conversations effortlessly. With support for WhatsApp, WhatsApp Business, Instagram, Facebook Messenger, and TikTok (coming soon), Herobot enables businesses of all sizes to provide instant responses and superior customer service at scale.
❤️ Support this project: If you find Herobot helpful, consider sponsoring the project on GitHub
Herobot is an open-source project that offers flexible deployment options to suit your needs:
-
Herobot Cloud (herobot.id): Coming soon! Our managed cloud solution at herobot.id will provide a hassle-free setup with automatic updates and maintenance. Perfect for businesses that want to get started quickly without infrastructure management.
-
Self-Hosting: Deploy Herobot on your own infrastructure for complete control and customization. Follow our setup guide below to host it on your servers.
Both options provide the same powerful features, letting you choose the deployment that best fits your requirements and privacy needs.
- Multi-Channel Support: Manage all your customer conversations from a single platform across multiple messaging channels
- Smart Business Tools: Seamlessly integrate with your existing tools - from shipping cost checks to Google Forms, spreadsheets, and custom API integrations
- Instant Responses: Provide 24/7 customer support, qualify lead automatically, and handle routine inquiries while your team focuses on high-value conversations
- Scalable Solution: Perfect for both small businesses and large enterprises, with the ability to manage multiple channels and teams from one dashboard
For a quick and easy setup without local installation, try these cloud-based options:
To set up the Herobot App locally, follow these steps:
-
Clone the Repository:
git clone [email protected]:herobot-id/herobot.git
-
Set Up Environment Variables:
cp .env.example .env
Variable Value Notes CHAT_SERVICE
gemini / openai / openrouter
Set to gemini for Google Gemini chat (free tier). but if you have OpenAI, you can use OpenAI as well, or if you want to use openrouter, you can choose it as well. EMBEDDING_SERVICE
gemini / openai
Set to gemini for embedding generation (free tier). but if you have OpenAI, you can use OpenAI as well. (openrouter doesn't have embedding models) GEMINI_API_KEY
<your-api-key>
Get an API-key at https://aistudio.google.com/apikey. OPENAI_API_KEY
<your-api-key>
Get an API-key at https://platform.openai.com/account/api-keys OPENROUTER_API_KEY
<your-api-key>
Get an API-key at https://openrouter.ai/settings/keys GEMINI_MODEL
gemini-2.0-flash-lite
(Optional but Better at RPM) OPENAI_MODEL
gpt-4o-mini
(Optional but Better at RPM) OPENROUTER_MODEL
meta-llama/llama-3.1-8b-instruct:free
(Recommended for free tier) -
Start Services
# Start all services docker compose up
This single command will automatically:
- Install Composer dependencies
- Install NPM dependencies
- Start the Laravel application
- Start the Vite development server
- Start the Reverb WebSocket server
- Start the WhatsApp server
-
Access the Application:
- The application will be accessible on port 80. Open your browser and navigate to
http://localhost
.
- The application will be accessible on port 80. Open your browser and navigate to
-
Credentials
Default user credentials are provided below:
Variable Value Email
[email protected]
Password
password
-
Stopping Services:
# Stop all Docker containers and services docker compose down
This will automatically stop all services including the Laravel application, Vite development server, Reverb WebSocket server, and WhatsApp server.
By following these steps, you will have the Herobot App up and running on your local machine, ready for development and testing.
We welcome contributions from the community! Whether you're fixing bugs, adding new features, or improving documentation, your help is appreciated.
Please read our Contributing Guide for detailed information on:
- Development setup and workflow
- Code standards and best practices
- Testing requirements
- Pull request process
- Channel integration guidelines
For bug reports and feature requests, please open an issue on GitHub.