BirdNet-Pi MCP Server
A Python-based Model Context Protocol (MCP) server for BirdNet-Pi integration.
Features
- Bird detection data retrieval with date and species filtering
- Detection statistics and analysis
- Audio recording access
- Daily activity patterns
- Report generation
Requirements
- Python 3.8+
- FastAPI
- Uvicorn
- Other dependencies listed in `requirements.txt`
Installation
- Clone the repository:
- Create a virtual environment and activate it:
- Install dependencies:
- Set up your data directories:
Configuration
The server can be configured using environment variables:
- `BIRDNET_DETECTIONS_FILE`: Path to detections JSON file (default: 'data/detections.json')
- `BIRDNET_AUDIO_DIR`: Path to audio files directory (default: 'data/audio')
- `BIRDNET_REPORT_DIR`: Path to reports directory (default: 'data/reports')
Running the Server
Start the server:
The server will run on `http://localhost:8000`.
API Endpoints
- `/functions` - List available functions (GET)
- `/invoke` - Invoke a function (POST)
Available Functions
- `getBirdDetections` - Get bird detections filtered by date range and species - Parameters: startDate, endDate, species (optional)
- `getDetectionStats` - Get detection statistics for a time period - Parameters: period ('day', 'week', 'month', 'all'), minConfidence (optional)
- `getAudioRecording` - Get audio recording for a detection - Parameters: filename, format ('base64' or 'buffer')
- `getDailyActivity` - Get bird activity patterns for a specific day - Parameters: date, species (optional)
- `generateDetectionReport` - Generate a report of detections - Parameters: startDate, endDate, format ('html' or 'json')