SMITHERY_DEPLOYMENT.mdโข3.9 kB
# Smithery Deployment Guide for Accounting Practice MCP
## ๐ Quick Deployment
This MCP server is configured for easy deployment on Smithery with the included `smithery.yaml` configuration file.
### Prerequisites
- Smithery account with deployment permissions
- Repository access to this codebase
- Docker (for local testing)
### Deployment Steps
1. **Verify Configuration**
```bash
# Check that all required files are present
ls -la Dockerfile smithery.yaml requirements.txt server/main.py
```
2. **Deploy to Smithery**
- Connect your repository to Smithery
- The `smithery.yaml` file will be automatically detected
- Smithery will build and deploy the MCP server
3. **Verify Deployment**
```bash
# Run the deployment health check
python deploy.py
```
## ๐ Configuration Details
### Dockerfile Configuration
The Dockerfile provides:
- **Base Image**: Python 3.11 slim for optimal size
- **Dependencies**: System packages (gcc, g++) and Python libraries
- **Environment**: Proper Python path configuration
- **Directories**: Creates necessary data directories
- **Command**: Runs the MCP server via stdio
### smithery.yaml Configuration
The deployment configuration includes:
- **Runtime**: Container-based execution
- **Build**: Uses Dockerfile for containerization
- **Start Command**: stdio mode for MCP protocol
- **Build Path**: Root directory (.)
### Available MCP Tools
1. `get_client_info` - Retrieve client information
2. `update_client_profile` - Update client profiles
3. `get_client_deadlines` - Get tax deadlines
4. `process_bank_statement` - Process bank statements
5. `reconcile_accounts` - Bank reconciliation
6. `calculate_tax_liability` - Tax calculations
7. `optimize_deductions` - Deduction optimization
8. `calculate_payroll` - Payroll processing
9. `sales_tax_calculation` - Sales tax calculations
10. `quickbooks_sync` - QuickBooks integration
## ๐ง Environment Variables
The following environment variables are configured:
- `PYTHONPATH`: Set to current directory
- `MCP_SERVER_NAME`: accounting-practice-mcp
- `MCP_SERVER_VERSION`: 1.0.0
## ๐ Monitoring
- **Health Check**: Available at `/health` endpoint
- **Metrics**: Request count, response time, error rate
- **Logs**: Info level with 7-day retention
- **Backup**: Daily backups with 30-day retention
## ๐ ๏ธ Troubleshooting
### Common Issues
1. **Build Failures**
- Verify Python 3.11 compatibility
- Check requirements.txt dependencies
- Ensure all source files are present
2. **Runtime Errors**
- Check health endpoint: `/health`
- Review logs for specific error messages
- Verify database initialization
3. **Tool Execution Issues**
- Ensure client data is properly initialized
- Check file permissions for data directories
- Verify MCP client connection
### Health Check
Run the health check script to verify deployment:
```bash
python deploy.py
```
Expected output:
```
๐ Initializing Accounting Practice MCP Server for Smithery deployment...
โ
Created directory: server/data/client_profiles
โ
Created directory: server/data/tax_tables
๐ Health Status: healthy
โ
Server is ready for deployment!
```
## ๐ Scaling
The server is configured for single replica deployment. For production scaling:
1. Update `smithery.yaml` deployment section
2. Enable auto-scaling if needed
3. Adjust resource requirements
4. Configure load balancing
## ๐ Security
- Encryption enabled (AES-256)
- Authentication disabled by default
- Client data isolation via separate databases
- Audit logging enabled
## ๐ Support
For deployment issues:
1. Check Smithery documentation
2. Review server logs
3. Run health check script
4. Verify configuration files
---
**Ready to deploy!** ๐
The server is fully configured and ready for Smithery deployment with comprehensive accounting automation capabilities.