Catalyst is an AI-powered multi-agent marketing platform designed to help Small and Medium Businesses (SMBs) scale their presales efforts through automated content creation, market research, campaign planning, and outreach.
- Market Research: Automatically gather market intelligence, trends, and competitor information
- Content Generation: Create text and visual content for multiple marketing channels
- Localization: Translate content to multiple languages with context preservation
- Campaign Planning: Develop data-driven marketing strategies
- Scheduling & Outreach: Automate content publishing and customer communication
- Backend:
- Python 3.8+
- FastAPI
- Uvicorn
- Asyncio
- Httpx
- Pydantic
- Frontend:
- Node.js 14+
- React 17+
- TypeScript
- Mantine UI
- React Router
Before setting up the application, you'll need to obtain API keys for the following services:
-
Apify API Key (for market research)
- Sign up at Apify
- Navigate to Account Settings > Integrations
- Create a new API key
-
DeepL API Key (for translation)
- Sign up at DeepL API
- Choose a plan (Free tier available)
- Copy your API key from the account dashboard
-
Vizcom API Key (for visual content generation)
- Sign up at Vizcom AI
- Request API access
- Copy your API key from the developer dashboard
-
Clone the repository:
git clone https://github.com/yourusername/catalyst-platform.git cd catalyst-platform -
Create a virtual environment:
python -m venv venv source venv/bin/activate # On Windows: venv\Scripts\activate
-
Install dependencies:
pip install -r requirements.txt
-
Set environment variables:
# Linux/macOS export APIFY_API_KEY=your_apify_api_key export DEEPL_API_KEY=your_deepl_api_key export VIZCOM_API_KEY=your_vizcom_api_key # Windows set APIFY_API_KEY=your_apify_api_key set DEEPL_API_KEY=your_deepl_api_key set VIZCOM_API_KEY=your_vizcom_api_key
Alternatively, create a
.envfile in the root directory:APIFY_API_KEY=your_apify_api_key DEEPL_API_KEY=your_deepl_api_key VIZCOM_API_KEY=your_vizcom_api_key -
Run the FastAPI server:
uvicorn backend.main:app --reload --host 0.0.0.0 --port 8000
-
Navigate to the frontend directory:
cd frontend -
Install dependencies:
npm install
-
Create a
.envfile with the backend API URL:REACT_APP_API_URL=http://localhost:8000 -
Start the development server:
npm start
-
The application should now be running at
http://localhost:3000
- Use the default credentials:
- Username:
admin - Password:
password
- Username:
The dashboard provides an overview of:
- Active campaigns
- Pending and completed content
- Agent status
- Recent activity
- Click "Create Content" on the dashboard or navigate to Content > New Content
- Fill in the content details:
- Select content type (social post, email, ad, etc.)
- Enter title and description
- Provide product information
- Set content parameters:
- Text options (tone, template, keywords)
- Visual options (style, aspect ratio, colors)
- Localization settings (target languages)
- Generate the content
- Review and approve the generated content
- Navigate to Campaigns > New Campaign
- Define campaign details:
- Name and description
- Target audience
- Start and end dates
- Goals and budget
- The system will automatically:
- Research market trends
- Identify target demographics
- Suggest content strategy
- Generate initial content
- Review and approve the campaign plan
- Monitor campaign progress on the dashboard
catalyst-platform/
├── backend/
│ ├── agents/
│ │ ├── market_research_agent.py
│ │ ├── content_generation_agent.py
│ │ ├── localization_agent.py
│ │ └── ...
│ ├── protocols/
│ │ └── mcp.py
│ └── main.py
├── frontend/
│ ├── public/
│ └── src/
│ ├── components/
│ │ ├── Dashboard.tsx
│ │ ├── ContentCreator.tsx
│ │ └── ...
│ └── App.tsx
├── requirements.txt
└── README.md
- Verify that your API keys are correctly set in environment variables
- Check that the backend server is running and accessible
- Ensure your firewall isn't blocking connections
- Check the logs for specific error messages
- Verify that the required API services (Apify, DeepL, Vizcom) are operational
- Restart the backend server to reinitialize agents
- Ensure you've provided sufficient product information
- Try different parameters (tone, style, etc.)
- For visual content issues, check that image URLs or sketches are valid
MIT
This project was created for the AI Agent Hackathon and utilizes the following sponsor tools:
For questions or support, please contact your-email@example.com