Smart Home -> Devices, then scroll to the bottom of that screen and select Forget All, then Discover. Works great now! Thanks for your help!

@F4u5t, here's an example to watch a folder for new files, and call the IFTTT webhook. $watcher = New-Object System.IO.FileSystemWatcher $watcher.Path = 'D:\folder_to_watch' $watcher.IncludeSubdirectories = $true $watcher.EnableRaisingEvents = $true $action = { $path = $event.SourceEventArgs.FullPath $changetype = $event.SourceEventArgs.ChangeType Write-Host "$path was $changetype at $(get-date)" $postdata = @{ value1=$path value2=$changetype } # To get your key, click the Documentation link at https://ifttt.com/maker_webhooks/ $key="your_ifttt_key" $event="new_file" $json = $postdata | ConvertTo-Json $response = Invoke-RestMethod "https://maker.ifttt.com/trigger/$event/with/key/$key" -Method Post -Body $json -ContentType 'application/json' write-host $response } Register-ObjectEvent $watcher 'Created' -Action $action "Waiting for events. Press CTRL-C to quit." while($true) { Start-Sleep .5 }

Obviously I can not run GUI programs as root. Background triggers work fine and do the job anyway. Thank you

Awesome @Rafael-Requena. Thanks for letting me know it's working.

@Russ It appears the service doesn't want to startup properly now. I have to manually start it up myself. Currently it's set as Automatic. and abort sleep isn't being discovered by my alexa.

Awesome @Eduardo-Lezcano, that's good progress. There are two reasons the voice commands wouldn't show up on the website: The voice field for the computer is empty. Use the Edit button on the website to fill that in. The voice field for the command is empty. Use the commands.json to fill that in ("voice":"firefox"). Can you verify both of them have a word in them? For a Raspberry Pi I like to use the word "pie" in the computer's voice field because it's what Alexa or Google Assistant think you're saying when you say Pi.

Awesome, got it working. Thank you.

@Byron-Trantham, sorry about the extra work but I'm glad you have a solution now.

@jbeckmantx, using the word ON with the normal google action ( as opposed to the newer TRIGFERcmd Smart Home action) has often been a problem because the word ON is used as the delimiter between the trigger name and the computer name (calculator on downstairs). I've tried to train Google's AI to handle that but it's still an intermittent issue. If you haven't tried the TRIGGERcmd Smart Home action, you really should. That's the best solution for thing you turn on and off.

it's ok after change tokken

Revised and re-posted here.

@dougcolt, sorry for the trouble, but you seem to have figured it out. Both computers and commands have a "voice" field. Both voice fields need to be populated, except for your default computer's voice field (although it doesn't hurt to populate it). By default, if you have one computer, that computer is your default computer.

@carlinhos, I see you haven't filled in the computer's voice field for the two laptops. The TRIGGERcmd Smart Home skill will only create devices if your computer and command have the voice field filled in. The exception to that rule is your default computer because your default computer's commands get Alexa device names that don't include the computer name (ie just "calculator" instead of "calculator on laptop").

@Russ Thank you very much

@Ryan694, my code is different from yours, but after looking here, I think this might work: const ExitHandler = { canHandle(handlerInput) { const request = handlerInput.requestEnvelope.request; return request.type === 'IntentRequest' && (request.intent.name === 'AMAZON.StopIntent'); }, handle(handlerInput) { return handlerInput.responseBuilder .speak('bye!') .shouldEndSession(true) .getResponse(); }, };

@killwa91, have you tried specifying the full path to your scripts or executables? So instead of your command being: myscript.sh Make it: /usr/local/bin/myscript.sh Also, about the forum having a separate account - I didn't write the forum software. I just installed and configured it, so it would be very difficult to do that integration.

@Josiah-Richards, yes, please do add more clarity. I think the bookmark URL's might solve part of this at least, but how do you see it working? Should there be a way to issue a special token with access to run specific trigger names on specific computers? Then you'd give that token to your external users / customers?