# AWS IoT Core [Health: Active]

**Category:** 🏭 Industrial & IoT  
**Repository:** https://github.com/nagarjunr/iot-mcp-servers  
**GitHub Stars:** 0  
**Views:** 3  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/aws-iot-core

## Description
MCP server for AWS IoT Core: device registry, shadows, jobs, rules, messaging.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `uvx` (confidence: high):

```json
"mcpServers": {
  "aws-iot-core": {
    "command": "uvx",
    "args": ["eclipse-ditto-mcp-server"]
  }
}
```

## Documentation

## What AWS IoT Core MCP server does

The AWS IoT Core MCP server connects an MCP-compatible AI client with AWS IoT Core. Its 48 tools cover the platform areas listed by the project: device registry, shadows, jobs, rules, and messaging. This gives an agent a structured way to work with AWS IoT Core instead of relying only on free-form instructions or external scripts.

The repository describes its IoT servers as providing read and write access to device registries, digital twins, and telemetry. For AWS IoT Core specifically, the supported resource areas are the registry, shadows, jobs, rules, and messaging.

## How it works

Each connector in the repository is an independent Python server. The AWS implementation has its own project configuration, lockfile, Dockerfile, MCP client configuration example, and server entry point rather than depending on a shared library layer.

The server communicates with AWS IoT Core through the platform-specific connector. It is not presented as a cross-vendor abstraction: the repository explicitly treats AWS IoT Core as a vendor-specific service. The project tests this connector against moto because AWS IoT Core does not have a free local emulator. That testing choice is useful context when evaluating behavior that depends on the complete AWS service rather than the emulator.

## Setup and configuration

The project requires Python 3.12 or newer and uses uv for dependency management and execution. The published package can be launched without cloning the repository:

```bash
uvx aws-iot-core-mcp-server
```

From a checkout, enter the `aws-iot-core` directory, install dependencies with `uv sync`, and start the server with the directory's Python entry point. The server reads its connection details and credentials from environment variables.

An `mcp.json` example is included in the server directory. Copy its relevant configuration into an MCP client's settings and fill in the required environment variables. The repository specifically identifies Claude Desktop and VS Code as supported configuration targets, while also describing the setup as usable with any MCP client.

## Tools and capabilities

The documented capability groups are:

- Device registry access
- Device shadow operations
- IoT jobs
- IoT rules
- IoT messaging

The server advertises 48 tools across these areas. The material does not specify which tools are read-only, which mutate resources, or whether individual operations require confirmation. Review the AWS IoT Core server README for those distinctions before granting it production credentials.

## Limitations and notes

AWS IoT Core is inherently tied to AWS, so this connector does not provide the repository's broader cross-vendor abstraction described for open protocols. Local testing uses moto rather than a free, complete AWS IoT Core instance; the repository notes that LocalStack IoT support requires a paid plan.

_Full upstream README: https://allmcps.com/mcp/aws-iot-core/readme_