---
title: "Get Containers Health"
method: GET
path: "/v1/mcp-server-instances/health/containers"
tags: ["v1", "protected", "mcp-server-instances"]
---

# Get Containers Health

`GET /v1/mcp-server-instances/health/containers`

Health of this workspace's MCP workloads.

The manager also has a route that answers for every workload it runs, which is
the wrong thing to hand a caller: its rows carry service names and container
ids belonging to other workspaces. The instance list comes from the
workspace-scoped service instead, and only those ids are asked about, so the
answer cannot name a workload the caller is not entitled to see.

A workload the manager reports as unhealthy, has never heard of, or cannot be
reached about is a fact about that instance, not a failure of the request: each
is reported per instance and the endpoint still answers 200.

## Response `200`

Successful Response

- MCPContainersHealthResponse
  - `healthy` integer, required
  - `instances` MCPInstanceHealthResponse[], required
    - `healthy` boolean, required
    - `instance_id` string, required
    - `name` string, nullable
    - `status` string, required
  - `total` integer, required

## Changes

- **2026-08-25** `ad32d7e4805a` — 1 breaking, 3 info
  - the response's body type changed from no type to `object` for status `200`
  - added the required property `healthy` to the response with the `200` status
  - added the required property `instances` to the response with the `200` status
  - added the required property `total` to the response with the `200` status

[Change history](https://skmtc.dev/agentarea/apis/agentarea-api/changes/v1/mcp-server-instances/health/containers/get.md)

---

[API](https://skmtc.dev/agentarea/apis/agentarea-api.md) · [All operations](https://skmtc.dev/agentarea/apis/agentarea-api/llms.txt) · [OpenAPI document](https://skmtc-service-production.skmtc.workers.dev/v1/apis/agentarea/agentarea-api/revisions/bc2efdac6eb4/schema)