Pkgx

can your ai run anything? yes it can

typescript

22

GitHub Stars

typescript

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": {
    "pkgxdev-mcp": {
      "command": "npx",
      "args": [
        "ts-node",
        "/path/to/pkgx-mcp/main.ts"
      ]
    }
  }
}

You can run Open Source tools through the pkgx MCP server to obtain actionable results and statistics, using a client that understands the Model Context Protocol. This guide walks you through using the server, installing it, and configuring it for practical work with an MCP client.

How to use

You consume this server from an MCP client to run tools via pkgx and gather outputs such as repository statistics or code analysis. Start by configuring a client to connect to the MCP server, then choose the tool you want to execute through the server. You will provide prompts or tasks to the MCP client, which delegates execution to pkgx and returns results you can use in your workflow.

How to install

Prerequisites: ensure you have Node.js installed on your system. You may also need pnpm or npm depending on how you install local dependencies.

Install pkgx if it is not already present on your system. You can rely on the installation flow you use for Open Source tooling, or obtain it through the standard install method provided by pkgx.
Clone the MCP server repository for pkgx MCP.
Set up the server configuration in your MCP client using the provided config snippet that points to the local runtime command for the server. The example is shown in the configuration block below.

Additional configuration and usage

Configuration examples show how to point an MCP client at the pkgx MCP server for local execution via a script or through npx, enabling you to run tools through the server. Use the stdio configuration that specifies the command and arguments to run the server locally. There are two viable configuration approaches demonstrated: one using a direct path to the main script, and another using npx with ts-node to execute the main script.

Security and cautions

Treat any tool execution through the MCP server with care. Ensure you run only trusted tools and limit access to sensitive files when possible. The server is designed to execute external tools via a controlled interface, but you should review the permissions and the scope of what each tool can access in your environment.

Examples you can use

Use pkgx to run a sequence of commands against a repository and generate statistics about the repository, such as a commit history, author activity, or file changes over time.

Troubleshooting tips

If you encounter issues connecting to the MCP server, verify the client configuration points to the correct stdio command and that the necessary script paths are accessible. Check that the required Node.js runtime and ts-node are available when using the npx approach. Review any error messages from the client for hints about missing dependencies or permission problems.

Available tools

Inspector

MCP Inspector to evaluate and debug an MCP server by executing ts-node on the main script to inspect its behavior and output.

PkgxRunner

Core capability to route tool execution through pkgx, allowing a client to run Open Source tools via the MCP server and receive structured results.