About mcp-daemonize
mcp-daemonize is an MCP server for AI agents managing long-running processes such as development servers. It provides controls to start and stop daemons, list them, and inspect their logs.
Key features
- 01Start long-running processes
- 02Stop daemons by name
- 03List running daemons
- 04Read recent daemon logs
Use cases
- 01Run Vite or Next.js servers from an agent
- 02Inspect server output while debugging
- 03Manage background processes in development workflows
MCP support
It exposes tools for MCP clients to start, stop, list, and retrieve logs from daemons.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.
Details
From the project READMEmcp-daemonize
mcp-daemonize is a Model Context Protocol (MCP) server designed to provide seamless management of long-running daemons (such as development servers) for AI agents like Claude Code and Cline. It enables advanced automation and debugging capabilities by allowing AI agents to start, stop, and monitor daemons, as well as access their logs in real time.
Motivation
While AI agents can execute commands, they are not well-suited for managing long-running development servers (e.g., vite dev). This is because the agent typically waits for the command to finish, which is not suitable for servers that are meant to run continuously during development. As a workaround, users often start development servers in a separate shell, but this prevents the AI agent from directly accessing the server's logs, making autonomous debugging impossible.
mcp-daemonize solves these problems by providing:
- An interface to start daemons without waiting for their termination.
- Tools to stop running daemons.
- Tools to view real-time logs of running daemons.
With these features, mcp-daemonize enables AI agents to fully manage development servers, monitor their output, and perform automated debugging, greatly expanding the possibilities for autonomous coding.
Use Cases
- Starting and stopping development servers (e.g., Vite, Next.js, etc.) from an AI agent.
- Viewing and analyzing real-time logs of running daemons for debugging.
- Automating development workflows that require persistent background processes.
- Enabling AI agents to manage, monitor, and debug long-running processes autonomously.
Prerequisites
- Go 1.24.2 or later installed.
- (Optional) Docker, if you wish to run mcp-daemonize in a container.
Installation
Download from GitHub Releases
Visit the Releases page to download the latest version.
- Download the archive for your platform from the latest release page.
- Extract the archive and move the binary to a directory in your PATH.
wget https://github.com/mackee/mcp-daemonize/releases/download/vX.Y.Z/mcp-daemonize_X.Y.Z_Darwin_x86_64.tar.gz
tar -xzf mcp-daemonize_X.Y.Z_Darwin_x86_64.tar.gz
sudo mv mcp-daemonize /usr/local/bin/
Build from Source
You can install the latest version using go install (Go 1.24.2 or later):
go install github.com/mackee/mcp-daemonize/cmd/mcp-daemonize@latest
Usage with MCP Host (e.g., Claude Code, Cline)
Add the following configuration to your MCP host settings (e.g., mcp.json):
{
"servers": {
"daemonize": {
"command": "/path/to/mcp-daemonize",
"args": [],
"env": {}
}
}
}
Replace /path/to/mcp-daemonize with the actual path to the built binary.
Usage
mcp-daemonize provides the following tools for AI agents:
Tools
-
daemonize_start
- Start a long-running process (e.g., a development server) as a daemon.
- Parameters:
name(string, required): Name of the daemon.command(string[], optional): Command to run (e.g.,["npm", "run", "dev"]).workdir(string, required): Working directory for the daemon (absolute path).
-
daemonize_stop
- Stop a running daemon by name.
- Parameters:
name(string, required): Name of the daemon to stop.
-
daemonize_list
- List all currently running daemons.
- Parameters: None
-
daemonize_logs
- Retrieve the latest logs from a running daemon.
- Parameters:
name(string, required): Name of the daemon.tail(number, required): Number of lines to read from the end of the log.
Example Workflow
- Start a development server as a daemon using
daemonize_start. - Use
daemonize_logsto monitor the server's output in real time. - Stop the server with
daemonize_stopwhen finished.
License
This project is licensed under the terms of the MIT open source license. See LICENSE for details.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.