Build flexible autonomous AI agents effortlessly. Set up in 2 minutes, run in 5, deploy in 10.
- Python 3.11+
- Node.js 18+ (for dashboard)
- Poetry (
pip install poetry)
# Clone the repository
git clone https://github.com/axioma-ai-labs/nevron.git
cd nevron
# Install Python dependencies
make deps
# Install dashboard dependencies
make dashboard-deps# Start everything (Agent + API + Dashboard)
make dev-fullThis starts:
- Dashboard: http://localhost:5173
- API: http://localhost:8000
- Agent: Running in background
All configuration is done through the Dashboard UI - no .env file needed!
- Open the dashboard at http://localhost:5173
- Go to Settings page
- Configure your LLM provider and API key
- Set up agent personality and goals
- Enable integrations as needed
Configuration is saved to nevron_config.json and loaded automatically.
Nevron uses a decoupled architecture where the API and Agent run as independent processes:
┌─────────────────┐ ┌──────────────────┐ ┌─────────────────┐
│ Dashboard │────▶│ FastAPI │────▶│ Shared State │
│ (Svelte) │ │ (Port 8000) │ │ ./nevron_state │
│ Port 5173 │◀────│ + WebSocket │◀────│ │
└─────────────────┘ └──────────────────┘ └────────┬────────┘
│
nevron_config.json ─────────┤
│
▼
┌─────────────────┐
│ Agent Runner │
│ (Independent) │
└─────────────────┘
Benefits:
- Configure everything from the UI
- Restart API without stopping the agent
- Multiple dashboards can connect simultaneously
- Agent continues running if dashboard disconnects
- Agent status (running/paused/stopped)
- Start, pause, stop controls
- Real-time statistics
- Live event feed via WebSocket
- Agent cycle history (Plan → Execute → Learn → Remember)
- Detailed cycle inspection
- Performance metrics
- LLM Configuration - Provider, model, API key
- Agent Settings - Personality, goal, behavior
- Actions - Enable/disable available actions
- Integrations - Configure Twitter, Discord, Telegram, etc.
- MCP Servers - Add custom tool servers
| Command | What it does |
|---|---|
make dev-full |
Agent + API + Dashboard (recommended) |
make dev |
API + Dashboard only (no agent) |
make run-agent |
Agent process only |
make api |
API server only |
make dashboard |
Dashboard only |
# Development
make deps # Install Python dependencies
make dashboard-deps # Install dashboard dependencies
make format # Format code
make lint # Run linting
make test # Run tests
# Running
make run # Run agent (legacy mode)
make run-agent # Run agent process (decoupled)
make api # Run API server
make dashboard # Run dashboard
make dev # Run API + dashboard
make dev-full # Run agent + API + dashboard
# Docker
make docker-up # Start all services
make docker-down # Stop all services
make docker-logs # View logs| Provider | Models |
|---|---|
| OpenAI | gpt-4o, gpt-4o-mini, gpt-4-turbo, o1, o1-mini |
| Anthropic | claude-sonnet-4, claude-opus-4 |
| xAI | grok-3, grok-3-fast, grok-2 |
| DeepSeek | deepseek-chat, deepseek-reasoner |
| Qwen | qwen-max, qwen-plus, qwen-turbo |
| Venice | llama-3.3-70b, qwen3-235b |
Nevron supports various external services (configure in Settings):
- Social Media: Twitter, Discord, Telegram, Slack, WhatsApp
- Development: GitHub
- Storage: Google Drive
- Media: Spotify, YouTube
- Web3: Lens Protocol
- Search: Tavily, Perplexity
For detailed documentation, visit: https://axioma-ai-labs.github.io/nevron/
This project is licensed under the Nevron Public License (NPL). See the LICENSE file for details.
Made with love by Neurobro