HTTP API¶
Lango exposes an HTTP API for health monitoring, agent discovery, authentication, and chat interaction.
Server Configuration¶
Settings:
lango settings→ Server
{
"server": {
"host": "localhost",
"port": 18789,
"httpEnabled": true
}
}
| Key | Type | Default | Description |
|---|---|---|---|
server.host | string | localhost | Bind address for the HTTP server |
server.port | int | 18789 | Port number |
server.httpEnabled | bool | true | Enable the HTTP server |
Endpoints¶
Health Check¶
GET /health
Returns the server health status. Use this for monitoring, load balancer probes, and Docker health checks.
curl http://localhost:18789/health
Agent Card (A2A)¶
GET /.well-known/agent.json
Returns the agent's A2A agent card when the A2A protocol is enabled (a2a.enabled: true). This endpoint follows the Agent-to-Agent protocol specification for remote agent discovery.
Authentication¶
Authentication endpoints are available when OIDC is configured. See Authentication for details on the OAuth login flow and token management.
Chat¶
The main chat endpoint accepts user messages and returns agent responses. When WebSocket is enabled, responses are streamed in real time via WebSocket events alongside the standard HTTP response.
Related¶
- WebSocket -- Real-time streaming events
- Authentication -- OIDC and OAuth configuration
- A2A Protocol -- Agent-to-Agent discovery