list_commandsLABSCO SUMMARY
An agent can press a button on a machine that is not in front of you.
The commands you already set up — a script, a reboot, the office lights — become things you ask for by name, with the parameter said in words. Describing a command in TRIGGERcmd turns it into its own tool, which is what makes the model pick the right one.
What it is
A local stdio server in front of your TRIGGERcmd account. It lists the commands registered across your computers and runs them, and it builds a dedicated tool for every command that carries an mcpToolDescription.
What you get
- Every TRIGGERcmd command on the account, across every computer
- Any of them run by naming the computer and the command
- Parameters passed through to commands that accept them
- One tool per described command, named run_{computer}_{command}, carrying that command's own description
Requirements
A TRIGGERcmd account and its token. The server takes the token from the environment when one is set there, otherwise it reads ~/.TRIGGERcmdData/token.tkn from your home directory. Adding an mcpToolDescription to a command takes effect after a server restart.
Setup effort
Run a container — build the image from the repository, then point your client at it
USE CASES
01
Running something on the desktop while you are on the laptop
List what is registered, then run the one you want by naming its computer and command.
02
Remembering what you set up months ago
Read every command on the account back, grouped by the machine it belongs to, without opening the TRIGGERcmd site.
03
Passing an argument without remembering its syntax
Say what you want in words — a colour, a filename, on or off — and it goes through as the command's parameter.
04
Giving the assistant one clear tool per job
Set an mcpToolDescription on the commands you use most; after a restart each appears as its own named tool, so the model stops guessing at the generic call.
2 tools
Listed in project docs. Your client may show a different list.
run_commandBEFORE YOU INSTALL
What to know first
Tools you get
2 tools named in the project's documentation.
Requires
- docker on PATH
- TRIGGERCMD_TOKEN in your environment
Where it runs
Both documented clients run the server on your machine.
INSTALLATION
Connect your client
Build the image from the repository, then point your client at it.
Server setup
TERMINAL
git clone https://github.com/rvmey/triggercmd-mcp-stdio
cd triggercmd-mcp-stdio
docker build -t triggercmd-mcp:latest .Then point your client at the image — the blocks under MCP Clients run triggercmd-mcp:latest with your TRIGGERcmd token
MCP Clients
Claude Desktop — mounted token file
CLAUDE DESKTOP — MOUNTED TOKEN FILE · JSON
{
"mcpServers": {
"triggercmd": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-v",
"/home/<YOUR_USERNAME>/.TRIGGERcmdData/token.tkn:/home/triggercmd/.TRIGGERcmdData/token.tkn:ro",
"triggercmd-mcp:latest"
]
}
}
}/home/<YOUR_USERNAME> is your own home directory — the part before the colon is the token file on your machine. Leave /home/triggercmd/... after the colon as it is: that is where the container expects the token. The container runs as UID 1000, so the file must be readable by that user.
VS Code — environment variable
VS CODE — ENVIRONMENT VARIABLE · JSON
{
"servers": {
"triggercmd": {
"type": "stdio",
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"-e",
"TRIGGERCMD_TOKEN=<YOUR_TRIGGERCMD_TOKEN>",
"triggercmd-mcp:latest"
]
}
}
}
