Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
coder-ka avatar

Procm

  • 4 repo stars
  • Updated May 31, 2026
  • coder-ka/procm-mcp

Procm is an MCP server that lets agents manage local OS processes through the Model Context Protocol.

About

Procm (procm-mcp) is a Model Context Protocol server dedicated to process management so AI coding agents can list, start, stop, or supervise OS processes in a structured way. Developers who constantly restart dev servers, queue workers, or one-off migration scripts benefit when the agent can manage those processes without fragile raw shell one-liners. Install the procm-mcp npm package and register it as a stdio MCP server in Claude Code, Cursor, or another compatible client. Skillselion places it under Operate → infra as a local runtime integration rather than a test runner or a library. Complexity is beginner to intermediate depending on how aggressively you let agents touch production-adjacent processes on your machine. Treat it as operational leverage: faster local iteration, not a replacement for proper production orchestration on Kubernetes or managed platforms.

  • MCP server focused on process management for agent-driven dev and ops tasks
  • npm package procm-mcp v0.0.43 with stdio transport
  • Title Procm in server metadata—lightweight supervision from the agent session
  • Fits solo workflows where one person juggles API, frontend, and worker processes locally
  • Complements shell skills by giving structured MCP tools instead of ad-hoc pkill patterns

Procm by the numbers

  • Data as of Jul 14, 2026 (Skillselion catalog sync)
terminal
claude mcp add procm-mcp -- npx -y procm-mcp

Add your badge

Show developers this MCP server is listed on Skillselion. Paste this into your README.

Listed on Skillselion
repo stars4
Packageprocm-mcp
TransportSTDIO
AuthNone
Last updatedMay 31, 2026
Repositorycoder-ka/procm-mcp

What it does

Expose local process start, stop, and supervision to your coding agent so dev servers, workers, and scripts are managed through MCP instead of manual terminals.

Who is it for?

Best when you run multiple local services and want the agent to start or stop them predictably during develop-and-debug loops.

Skip if: Production cluster orchestration, untrusted multi-tenant hosts, or teams that forbid agents from touching OS processes.

What you get

Once Procm is registered, your agent can use MCP process-management tools to supervise the processes you allow on that host.

  • Registered Procm stdio MCP server
  • Agent-accessible process management on the configured host
  • More repeatable local start/stop workflows during development

By the numbers

  • Server version 0.0.43 in server.json
  • npm identifier procm-mcp with stdio transport
  • Published title Procm in MCP metadata
README.md

procm-mcp

A Model Context Protocol (MCP) server for process management.

Supported features

  • Secure and automatable process creation
  • Cleanup created processes automatically on termination (e.g. exiting claude code)
  • Common process management features supported, restarting, deleting, checking status or retreving stdout/stderr of processes

Using these features, LLMs start processes like development servers, docker-compose, or test watchers and check their outputs to fix bugs automatically.

Installation

npm i -D procm-mcp

.mcp.json

{
  "mcpServers": {
    "procm-mcp": {
      "command": "node",
      "args": ["./node_modules/procm-mcp/build/index.js"],
      "env": {}
    }
  }
}

Secure process creation

You can permit LLMs to use start-process tool without confirmation, because procm-mcp only allow whitelisted process creations.

LLMs will ask you to use allow-start-process tool to add specific process creation to the whitelist.

Once you allow a process creation, you don't have to confirming it anymore as long as the command and the working directory are the same.

I call it "allow-x pattern", which can balances security and usability in MCP.

Warning: Do not permit LLMs to use allow-start-process without confirmation.That means "Do anything you want to".

Tools

  • allow-start-process Allow specific processes to be created
    • script (required): The script/command to allow
    • args (optional): Array of arguments
    • cwd (optional): Working directory
  • start-process Start a new process with specified script and arguments
    • script (required): The script/command to execute
    • name (optional): A friendly name for the process
    • args (optional): Array of arguments to pass to the script
    • cwd (required): Working directory for the process
    • envs (optional): Environment variables to set for the process
  • delete-process Stop and remove a process by ID.The default signal is SIGTERM, but SIGKILL(force killing) will be sent after 10 seconds unless the process exits.
    • id (required): The process ID
  • restart-process Restart an existing process by ID
    • id (required): The process ID
  • get-process-info Get detailed information about a process
    • id (required): The process ID
  • list-processes List all currently managed processes
    • No parameters required
  • get-process-stdout Retrieve stdout logs from a process
    • id (required): The process ID
    • chunkCount (optional): Number of recent log entries to retrieve (default: 10)
  • get-process-stderr Retrieve stderr logs from a process
    • id (required): The process ID
    • chunkCount (optional): Number of recent log entries to retrieve (default: 10)

License

MIT

Recommended MCP Servers

How it compares

Local process-management MCP server, not a full container platform or remote PaaS deploy integration.

FAQ

Who is Procm for?

Developers using MCP agents who need dependable local process supervision while building and operating small apps alone.

When should I use Procm?

Use it when your agent routinely starts dev servers, workers, or scripts and you want structured process tools instead of repeated manual terminal commands.

How do I add Procm to my agent?

Add a stdio MCP entry for the procm-mcp npm package in your agent configuration and restart the client on the machine where processes run.

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.