# Testkube MCP

> Testkube MCP is a freemium MCP server for developer tools. Run, inspect and troubleshoot Testkube test workflows from an AI assistant. It can run locally on your machine or as a hosted remote server.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Platforms:** API, self-hosted, Linux
- **Repository:** https://github.com/kubeshop/testkube
- **MCP Registry name:** io.github.kubeshop/testkube-mcp
- **Categories:** developer tools, cloud and infrastructure
- **Listing updated:** 2026-11-14

## About

Testkube orchestrates tests inside Kubernetes clusters, and this server lets an assistant reach that setup: the registry entry says it manages test workflows, executions and artifacts. It runs as a stdio container image from the kubeshop organisation or as a remote streamable-HTTP endpoint scoped to an organisation and environment on api.testkube.io.

Authentication uses a Testkube access token plus organisation and environment IDs, with optional overrides for the control plane and dashboard URLs. The wider platform runs any kind of test (API, end-to-end, performance, security, infrastructure) and triggers them manually, on schedules, from CI or GitOps pipelines, on Kubernetes events or through MCP.

The repository README describes an MIT-licensed open-source agent for a single cluster and a commercial control plane that adds Testkube AI for workflow generation, failure investigation and remediation pull requests.

## Features

- **Workflow management:** Create and inspect Testkube test workflows through an assistant.
- **Execution and artifact access:** Look up test runs, logs and artifacts for troubleshooting.
- **Stdio or remote:** Run the container locally or connect to the hosted per-environment endpoint.
- **Environment-scoped auth:** An access token with organisation and environment IDs limits what the server can see.
- **Fits the wider platform:** Works alongside Testkube triggers, webhooks and the REST API.

## Good for

- Ask why last night's end-to-end run failed
- Trigger a performance test workflow from a chat
- Pull artifacts from a failed execution into a debugging session
- Review which workflows exist in an environment

## Install

### Claude Code (Terminal)

```sh
claude mcp add testkube-mcp --env TK_ACCESS_TOKEN='<TK_ACCESS_TOKEN>' --env TK_ORG_ID='<TK_ORG_ID>' --env TK_ENV_ID='<TK_ENV_ID>' --env TK_MCP_SHTTP_AUTH_TOKEN='<TK_MCP_SHTTP_AUTH_TOKEN>' -- docker run -i --rm -e TK_ACCESS_TOKEN -e TK_ORG_ID -e TK_ENV_ID -e TK_MCP_SHTTP_AUTH_TOKEN docker.io/kubeshop/mcp-server:2.15.2
```

### Claude Desktop (claude_desktop_config.json)

```json
{
  "mcpServers": {
    "testkube-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "TK_ACCESS_TOKEN",
        "-e",
        "TK_ORG_ID",
        "-e",
        "TK_ENV_ID",
        "-e",
        "TK_MCP_SHTTP_AUTH_TOKEN",
        "docker.io/kubeshop/mcp-server:2.15.2"
      ],
      "env": {
        "TK_ACCESS_TOKEN": "<TK_ACCESS_TOKEN>",
        "TK_ORG_ID": "<TK_ORG_ID>",
        "TK_ENV_ID": "<TK_ENV_ID>",
        "TK_MCP_SHTTP_AUTH_TOKEN": "<TK_MCP_SHTTP_AUTH_TOKEN>"
      }
    }
  }
}
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "testkube-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "TK_ACCESS_TOKEN",
        "-e",
        "TK_ORG_ID",
        "-e",
        "TK_ENV_ID",
        "-e",
        "TK_MCP_SHTTP_AUTH_TOKEN",
        "docker.io/kubeshop/mcp-server:2.15.2"
      ],
      "env": {
        "TK_ACCESS_TOKEN": "<TK_ACCESS_TOKEN>",
        "TK_ORG_ID": "<TK_ORG_ID>",
        "TK_ENV_ID": "<TK_ENV_ID>",
        "TK_MCP_SHTTP_AUTH_TOKEN": "<TK_MCP_SHTTP_AUTH_TOKEN>"
      }
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "testkube-mcp": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "TK_ACCESS_TOKEN",
        "-e",
        "TK_ORG_ID",
        "-e",
        "TK_ENV_ID",
        "-e",
        "TK_MCP_SHTTP_AUTH_TOKEN",
        "docker.io/kubeshop/mcp-server:2.15.2"
      ],
      "env": {
        "TK_ACCESS_TOKEN": "<TK_ACCESS_TOKEN>",
        "TK_ORG_ID": "<TK_ORG_ID>",
        "TK_ENV_ID": "<TK_ENV_ID>",
        "TK_MCP_SHTTP_AUTH_TOKEN": "<TK_MCP_SHTTP_AUTH_TOKEN>"
      }
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "testkube-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "TK_ACCESS_TOKEN",
        "-e",
        "TK_ORG_ID",
        "-e",
        "TK_ENV_ID",
        "-e",
        "TK_MCP_SHTTP_AUTH_TOKEN",
        "docker.io/kubeshop/mcp-server:2.15.2"
      ],
      "env": {
        "TK_ACCESS_TOKEN": "<TK_ACCESS_TOKEN>",
        "TK_ORG_ID": "<TK_ORG_ID>",
        "TK_ENV_ID": "<TK_ENV_ID>",
        "TK_MCP_SHTTP_AUTH_TOKEN": "<TK_MCP_SHTTP_AUTH_TOKEN>"
      }
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "testkube-mcp": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "TK_ACCESS_TOKEN",
        "-e",
        "TK_ORG_ID",
        "-e",
        "TK_ENV_ID",
        "-e",
        "TK_MCP_SHTTP_AUTH_TOKEN",
        "docker.io/kubeshop/mcp-server:2.15.2"
      ],
      "env": {
        "TK_ACCESS_TOKEN": "<TK_ACCESS_TOKEN>",
        "TK_ORG_ID": "<TK_ORG_ID>",
        "TK_ENV_ID": "<TK_ENV_ID>",
        "TK_MCP_SHTTP_AUTH_TOKEN": "<TK_MCP_SHTTP_AUTH_TOKEN>"
      }
    }
  }
}
```

## FAQ

### What credentials does it need?

TK_ACCESS_TOKEN, TK_ORG_ID and TK_ENV_ID are required; the control plane URL defaults to https://api.testkube.io.

### Is there a remote endpoint?

Yes, a streamable-HTTP URL of the form api.testkube.io/organizations/{organization_id}/environments/{environment_id}/mcp.

### Is Testkube open source?

The README says the standalone agent is MIT-licensed; a commercial control plane adds multi-cluster dashboards, SSO and AI features.

### Which test types does Testkube run?

API, end-to-end, performance, security and infrastructure tests, among others.

### Is Testkube MCP free?

Testkube MCP has a free plan, with paid plans for more usage or features.

### How do I add Testkube MCP to Claude Code?

Run this command in your terminal: claude mcp add testkube-mcp --env TK_ACCESS_TOKEN='<TK_ACCESS_TOKEN>' --env TK_ORG_ID='<TK_ORG_ID>' --env TK_ENV_ID='<TK_ENV_ID>' --env TK_MCP_SHTTP_AUTH_TOKEN='<TK_MCP_SHTTP_AUTH_TOKEN>' -- docker run -i --rm -e TK_ACCESS_TOKEN -e TK_ORG_ID -e TK_ENV_ID -e TK_MCP_SHTTP_AUTH_TOKEN docker.io/kubeshop/mcp-server:2.15.2

### What are the best alternatives to Testkube MCP?

The closest alternatives to Testkube MCP are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Concord MCP](https://ai-powered.com/mcp/concord-mcp): Concord MCP is a freemium MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/testkube-mcp
