Steadybit MCP Server

Name: Steadybit MCP Server
Author: steadybit

@steadybit

代码仓库

关于 Steadybit MCP Server

Steadybit MCP-Server

#mcp #steadybit #server

基本信息

分类

其他

许可证

MIT license

运行时

java

传输方式

stdio

发布者

steadybit

配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcp-steadybit": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "java",
        "-jar",
        "target/mcp-1.0.0-SNAPSHOT.jara",
        "-e",
        "API_URL=https://platform.steadybit.com/api",
        "-e",
        "API_TOKEN=123456"
      ]
    }
  }
}

工具

List of experiment designs with their key and name

The experiment design

The experiment execution

List of actions

概览

What is Steadybit MCP Server?

Steadybit MCP Server is a Model Context Protocol server that enables LLM tools like Claude to interact with the Steadybit platform. It allows AI assistants to list, retrieve, and manage experiment designs and executions, as well as query registered actions.

How to use Steadybit MCP Server?

You need a Steadybit account and an API token. Set the API_TOKEN environment variable (required) and optionally API_URL (defaults to https://platform.steadybit.com/api). The server can be run via Docker using ghcr.io/steadybit/mcp:latest or built locally from source with Maven and launched directly with Java. Configure it in Claude Desktop under Settings → Developer by adding a JSON entry for the steadybit MCP server.

Key features of Steadybit MCP Server

List experiment designs for a given team
Retrieve a full experiment design by its key
Search and filter experiment executions with multiple criteria
Get details of a specific experiment execution by ID
List currently registered actions with pagination

Use cases of Steadybit MCP Server

An AI assistant queries a team’s experiment designs to plan a new experiment
A developer retrieves the full definition of an existing experiment design
A user filters experiment executions by state or date range to review results
A platform operator inspects a single execution to diagnose a failure
An engineer lists available actions to understand what experiments can be created

FAQ from Steadybit MCP Server

What is the Steadybit MCP Server used for?

It gives LLM tools like Claude the ability to read and interact with the Steadybit chaos engineering platform via standard MCP tools.

What prerequisites are needed?

A Steadybit account and an API token (Admin or Team level) from Settings → API Access Tokens.

How do I run the server?

You can run it with Docker (ghcr.io/steadybit/mcp:latest) or build the Java project with Maven and execute the JAR locally, passing API_TOKEN and optionally API_URL as environment variables.

Where are the server logs?

Logs are written to steadybit-mcp.log in the working directory, or to a custom path set via the LOGGING_FILE_NAME environment variable. No output is sent to stdout.