MCP This

Exposes CLI tools and AI prompt templates via YAML to MCP Clients.
  • javascript

14

GitHub Stars

javascript

Language

5 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": {
    "shane-kercheval-mcp-this": {
      "command": "uvx",
      "args": [
        "mcp-this",
        "--config-path",
        "/path/to/your-tools.yaml"
      ]
    }
  }
}

You can turn any CLI tool into an MCP tool and expose reusable prompt templates to MCP Clients like Claude Desktop. By configuring YAML files, you define the commands you want to run and the prompts you want to reuse, and the MCP Server makes them available to clients in a structured, easy-to-use way.

How to use

Set up an MCP Client session and connect to one of the available MCP servers described below. You can then run your custom CLI tools and invoke prompt templates from the client. Practical use patterns include: listing available tools, executing a specific tool with parameters, and prompting an AI client to apply a structured prompt with your defined arguments.

Common workflows you can perform: you can execute tools like get-current-time or system-info that you define in YAML, and you can use prompts such as code-reviewer or summarize-webpage to guide AI-assisted tasks. The server handles parameter substitution and executes the underlying commands, returning structured results that your client can display or pass to AI workflows.

To begin, configure a client to connect to the MCP server by pointing it at one of the provided runtime configurations. For example, you might provide the command runner and the path to your YAML configuration so that the client can discover and use the tools and prompts you defined.

How to install

Prerequisites: you need a runtime capable of launching MCP tooling, and you should have an MCP runner available (such as uvx) that can execute the configured commands in isolated environments.

Step-by-step setup for a typical local workflow:

# Install uvx (includes the MCP runtime if not already present)
# This is a typical one-line install shown in common setups
curl -LsSf https://astral.sh/uv/install.sh | sh

# Prepare your YAML configuration file with your tools and prompts
# Save as my-tools.yaml and customize paths and commands as needed

# Run the MCP server-configured tooling locally (stdio runtime)
# The exact command may vary; the snippet below demonstrates the expected pattern
uvx mcp-this --config-path ./my-tools.yaml

Configuration and usage notes

The MCP server config is defined in YAML or JSON blocks that describe tools and prompts. Tools execute shell commands with parameter placeholders, while prompts define Handlebars-style templates that the client can fill in with arguments.

Here are example runtime connections you can deploy in your client setup to access the MCP server:

{
  "mcpServers": {
    "mcp-this-custom": {
      "command": "uvx",
      "args": ["mcp-this", "--config-path", "/path/to/your-tools.yaml"]
    }
  }
}

Troubleshooting and notes

If you encounter issues connecting to the MCP server, verify that your YAML/JSON configuration paths exist and that the runtime command (such as uvx) is accessible on your system. Ensure the version of the runner supports the features described in your configuration and that any file paths used by tools and prompts are correct.

Examples you can start with

## Custom server configuration (example 1)
{
  "mcpServers": {
    "mcp_this_custom": {
      "command": "uvx",
      "args": ["mcp-this", "--config-path", "/path/to/your-tools.yaml"]
    }
  }
}
## Preset configuration (example 2)
{
  "mcpServers": {
    "mcp_this": {
      "command": "uvx",
      "args": ["mcp-this", "--preset", "default"]
    }
  }
}

Available tools

get-current-time

Display the current date and time in various formats.

system-info

Get basic system information such as kernel, OS, and CPU cores.

web-scraper

Fetch a webpage and convert it to clean, readable text.

find-large-files

Find files larger than a specified size within a directory.

git-status-summary

Provide a concise overview of git repository status.

test-runner

Run tests with an optional pattern to filter which tests to execute.

docker-container-logs

Retrieve logs from a Docker container.

port-checker

Check which process is using a specific port.

service-status

Check the status of a system service.

disk-usage-analyzer

Analyze disk usage and identify the largest directories.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational