API Reference
Admin Monitoring
System-wide monitoring endpoints for platform health, metrics, and audit logs.
Admin monitoring endpoints provide visibility into platform health, service metrics, and audit trails. All endpoints require ADMIN role and are scoped to the current tenant except for super-admin endpoints (internal use only).
Endpoints
| Method | Endpoint | Description |
|---|---|---|
| GET | /admin/metrics/system | CPU, memory, uptime metrics |
| GET | /admin/metrics/agents | Agent pipeline status (planner, interviewer, assessor) |
| GET | /admin/metrics/workflows | Workflow state counts across tenant |
| GET | /admin/dashboard/interviews?limit=N | Active interview records for dashboard |
GET /admin/metrics/system
Returns system-level metrics: CPU usage, memory, process uptime.
bash
curl https://mayaapi.teamcast.ai/api/v1/admin/metrics/system \
-H "Authorization: Bearer <admin-jwt>"Response 200
{
"cpu": { "usage": 12.4 },
"memory": { "used": 245, "total": 512, "unit": "MB" },
"uptime": 86400,
"timestamp": "2025-02-24T12:00:00.000Z"
}GET /admin/metrics/agents
Returns health and status of the three Agno Python agents.
bash
curl https://mayaapi.teamcast.ai/api/v1/admin/metrics/agents \
-H "Authorization: Bearer <admin-jwt>"Response 200
{
"planner": { "status": "up", "port": 7777, "latency": 12 },
"interviewer": { "status": "up", "port": 7778, "latency": 8 },
"assessor": { "status": "up", "port": 7779, "latency": 10 }
}GET /admin/metrics/workflows
Returns workflow state counts across all interviews in the tenant.
bash
curl https://mayaapi.teamcast.ai/api/v1/admin/metrics/workflows \
-H "Authorization: Bearer <admin-jwt>"Response 200
{
"RECEIVED": 0,
"INFO_NEEDED": 1,
"VALIDATING_SKILLS": 0,
"GENERATING_PLAN": 1,
"PENDING": 3,
"APPROVED": 12,
"IN_PROGRESS": 2,
"COMPLETED": 84,
"ASSESSMENT_PENDING": 2,
"ASSESSMENT_APPROVED": 78
}GET /admin/dashboard/interviews
Returns active interview records for the admin dashboard overview. Use the limit query parameter to control how many are returned.
bash
curl "https://mayaapi.teamcast.ai/api/v1/admin/dashboard/interviews?limit=20" \
-H "Authorization: Bearer <admin-jwt>"GET /admin/kafka/status
Returns consumer group lag for each Kafka topic. High lag indicates a processing backlog.
Response 200
{
"consumerGroups": [
{
"groupId": "api-gateway-consumer",
"topics": [
{ "topic": "interview.plan.created", "lag": 0 },
{ "topic": "interview.assessment.ready", "lag": 0 },
{ "topic": "interview.assessment.completed", "lag": 0 }
]
}
],
"brokerStatus": "CONNECTED",
"brokerUrl": "<kafka-broker-host>"
}For production deployments, expose these metrics to Prometheus via the
/api/v1/health/metrics endpoint and visualize with the provided Grafana dashboard templates.Was this page helpful?