- Home
- MCP servers
- Commander
Commander
- rust
1
GitHub Stars
rust
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": {
"towry-commander-mcp": {
"command": "/usr/local/bin/commander-mcp",
"args": []
}
}
}You run a dedicated MCP server to manage background processes with high performance, reliable logging, and clear per-process control. This server uses a standard MCP protocol to start, monitor, and control long-running tasks, giving you programmatic access to run, read logs, restart, and stop processes while keeping persistent logs for auditing and troubleshooting.
How to use
You connect to the MCP server via a client that speaks the MCP protocol. Use the available tools to start processes, monitor their status, read historical logs, and control lifecycle actions like restart or stop. All interactions are performed through the MCP client, and you can access per-process logs and runtime metrics to understand how each background task behaves.
How to install
Prerequisites: you need a Rust toolchain and the cargo build system.
# Install Rust toolchain if you haven’t already
# Visit https://www.rust-lang.org/tools/install for official guidance
# Build from source (release build recommended)
cargo build --release
# Run the server (in release mode)
cargo run --release
Configuration and usage notes
To run the MCP server locally, you can start the server as a standard IO (stdio) process. The following configurations illustrate how you can integrate with clients like Claude Desktop or test tooling.
{
"mcpServers": {
"commander": {
"type": "stdio",
"command": "/usr/local/bin/commander-mcp",
"args": []
}
}
}
Examples of starting and interacting with the server
Start the server from a development workflow or a packaged build, then connect a client to the standard input/output endpoint. Use the client to send commands to run, stop, or inspect processes and to fetch logs.
# If you’re running a prebuilt binary installed in /usr/local/bin
/usr/local/bin/commander-mcp
# If you’re running from a Cargo context (development)
cargo run --release
Available tools
run
Execute a command in the background. It waits a moment to detect early failures and returns a process ID with initial logs.
kill
Terminate a running process by its ID.
kill_all
Terminate all currently running processes.
read
Read the stdout logs for a specific process, returning recent output up to a configurable limit.
restart
Restart a previously started process by its ID.
list
List all processes with status, PID, start time, uptime, and resource usage.