1. Conduid
  2. Developer Tools
  3. Cmd
MCP server · Developer Tools

Cmd

by · github.com/developit/mcp-cmd

CLI for calling successive MCP Server tools

Unclaimed MIT last commit a year ago mcp-clientdevtoolsmcpmcp-tools
InstallView source
57Fair

Scored 5 months ago · breakdown

About Cmd

Cmd is an MCP server published by developit in the Developer Tools category: cLI for calling successive MCP Server tools.

The repository has 24 stars and 4 forks, with the last commit a year ago. Six months or more without a commit doesn't mean the server is broken, but check the open issues before depending on it in production.

Install

From source
git clone https://github.com/developit/mcp-cmd

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Cmd

Powered by Claude · Grounded in docs

I know everything about Cmd. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

Static checks derived from the repository yesterday. Heuristics on the source, not an audit.

  • ✓README presentThe repository ships documentation.
  • ✓License declaredUsable under a stated license.
  • !Tests presentNo tests found. Behaviour changes may go unnoticed between versions.
  • ✓Dependencies pinnedA lockfile pins dependency versions.
  • ✓No dynamic code executionSource scan found no eval, dynamic Function, or shell execution.
  • !Scoped permissionsDoesn't declare a permission scope. Assume it can do anything its process can.

README

mcp-cmd

A simple Node.js-based command line MCP Client that exposes commands for listing and calling MCP tools and prompts.

Why?

There are other command line MCP Clients out there, but all of them spawn the referenced MCP Servers for every tool call.

mcp-cmd is different: you use its CLI to start and stop MCP servers, and the servers run in the background. That means successive tool calls to an MCP Server are handled by the same running instance of that server.

Usage

First, start a server by giving it a name and the command + args, then interact with it, then stop it.

# Start an MCP server
mcp-cmd start <servername> <...url-or-commands>

# List server tools
mcp-cmd tools <servername>

# List server prompts
mcp-cmd prompts <servername>

# Call a tool provided by the server
mcp-cmd call <servername> <toolname> --arg1=value1 --arg2=value2

# Call a prompt provided by the server
mcp-cmd prompt <servername> <promptname> --arg1=value1 --arg2=value2

# Stop the MCP server
mcp-cmd stop <servername>

Example

# start server
mcp-cmd start puppeteer npx -y @modelcontextprotocol/server-puppeteer

# list tools
mcp-cmd tools puppeteer

# navigate to a page, log the title, then take a screenshot
mcp-cmd tools puppeteer navigate --url=https://example.com
mcp-cmd tools puppeteer execute --code=document.title
mcp-cmd tools puppeteer screenshot --name=example

# stop the server
mcp-cmd stop puppeteer

Implementation Details

  • dead simple CLI built using sade
  • based on the official @modelcontextprotocol/sdk package
  • only supports starting and stopping MCP servers, listing and calling tools (no resources)
  • stderr from the tool during the call is forwarded to stderr

README fetched from the repository yesterday. The original is authoritative.

Questions

About Cmd

How do I install Cmd?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Cmd safe to use with an AI agent?

Its trust score is 57 out of 100 (fair). It passes 4 of the 6 static security checks run so far; the failures are listed above. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Cmd still maintained?

The last commit was a year ago. That's long enough that you should check whether the maintainer is responding to issues before depending on it.