Service Deployment
Use AIGNE CLI to deploy agents as services.
aigne serve Command#
Serve agents in the specified directory as a Model Context Protocol (MCP) server.
aigne serve [path] [options]
Parameters#
path: Path to the agent directory (defaults to current directory.)
Options#
--mcp: Serve agents as an MCP server (currently required option)--host <host>: Host address to run the MCP server on (defaults to "localhost"), use "0.0.0.0" to expose the server publicly--port <port>: Port to run the MCP server on (uses PORT environment variable if set, otherwise defaults to 3000)--pathname <pathname>: URL path for the MCP server endpoint (defaults to "/mcp")--help: Show command help
Basic Usage Examples#
Serve MCP service on default port#
aigne serve --mcp
After successful startup, the command will display the server running address:
MCP server is running on http://localhost:3000/mcp
Serve in specific directory#
aigne serve ./my-agents --mcp
Use custom port#
aigne serve ./my-agents --mcp --port 8080
Expose server publicly#
aigne serve --mcp --host 0.0.0.0
Use custom endpoint path#
aigne serve --mcp --pathname /api/agents
Advanced Configuration Examples#
Complete custom configuration#
aigne serve ./my-agents --mcp --host 0.0.0.0 --port 8080 --pathname /api/agents
This will serve the MCP service at http://0.0.0.0:8080/api/agents.
Environment Variables#
You can use environment variables to configure the server:
# Set port
export PORT=8080
aigne serve --mcp
# Or set directly in command line
PORT=8080 aigne serve --mcp
MCP Server#
Model Context Protocol (MCP) is a standard protocol that allows AI models to interact with external tools and data sources. By serving your AIGNE agents as an MCP server, you can:
- Integrate into other applications: Allow other applications to use your agents through the MCP protocol
- Provide API access: Offer a standardized API interface for your agents
- Extend functionality: Allow multiple clients to access your agent services simultaneously
Deployment Considerations#
- Security: Ensure appropriate security measures when deploying in production environments
- Performance: Adjust server configuration based on expected load
- Monitoring: Implement proper logging and monitoring
- Network configuration: Ensure firewall and network settings allow required connections