Hopx

Provides isolated code execution in cloud containers with support for Python, JavaScript, Bash, and Go, plus data analysis and file operations.

python

172

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "hopx-ai-mcp": {
      "command": "uvx",
      "args": [
        "hopx-mcp"
      ],
      "env": {
        "HOPX_API_KEY": "YOUR_API_KEY",
        "HOPX_BASE_URL": "https://api.hopx.dev"
      }
    }
  }
}

The HOPX MCP Server lets you run code in isolated cloud containers, enabling AI assistants to execute Python, JavaScript, Bash, and Go with access to common data tools and file operations. It provides safe, ephemeral sandboxes that auto-clean up after use, making it practical for executing user-provided code and data processing tasks from your AI workflows.

How to use

You interact with the HOPX MCP Server through a client that supports the MCP protocol. Create or select a sandbox for your session, then run code or commands inside the sandbox. You can execute one-shot code, reuse a sandbox for multiple operations, and perform file I/O and shell commands. Each operation runs in an isolated container, and containers are cleaned up automatically after use to prevent resource leaks.

How to install

Prerequisites you need before starting:
- Node.js or a compatible runtime is not strictly required here since the MCP client is expected to invoke the MCP server via the uvx command, but you should have the uvx CLI installed to run the server endpoint.

Additional sections

Configuration notes, security considerations, examples, troubleshooting, and other important details are provided below to help you get started and keep your setup secure and reliable.

Available tools

create_sandbox

Create a new sandbox session to run code in an isolated container.

list_sandboxes

List all active or previously created sandboxes.

delete_sandbox

Terminate and clean up a sandbox by its ID.

execute_code_isolated

Run a one-shot piece of code inside a fresh sandbox and return stdout, exit code, and timing.

execute_code

Execute code within an existing sandbox identified by sandbox_id.

execute_code_background

Run long-running or background tasks inside a sandbox.

execute_code_async

Register a task with a webhook callback for extended runtimes.

file_read

Read a file from the sandbox filesystem.

file_write

Write data to a file inside the sandbox.

file_list

List files in a directory inside the sandbox.

file_exists

Check if a file or directory exists inside the sandbox.

file_remove

Remove a file inside the sandbox.

file_mkdir

Create a directory inside the sandbox.

run_command

Execute a shell command inside the sandbox.

run_command_background

Run a shell command in the background inside the sandbox.

env_get

Get an environment variable value from the sandbox context.

env_set

Set an environment variable inside the sandbox.

env_clear

Clear environment variables in the sandbox context.