You can run an automation from the canvas to test it, or run a placed copy from the Tools page, the Ribbon, a project folder, Power Strip, or a trigger. They all use the same runner, so an automation behaves the same way wherever you start it.
Canvas Runs
Run on the Automations tab tests the current canvas without publishing it. It checks the automation first and stops with a message if something is wrong, such as a missing Tool, a loop, a card with no path to an Output, or a conditional line without a rule.
A canvas run starts with empty input, and document cards on the canvas aren't used. Cards update as each step runs, and a run log appears at the end.
Deployed Runs
A placed automation uses the best input available, in this order:
- Input you supply when you start it, such as files dropped on a Power Strip button, a captured browser page, or the current selection when you start it from the Ribbon.
- Its saved Documents or folders.
- Empty input.
A project-folder placement always uses its folder as input.
Where you can start a placed automation:
- Tools page — select the automation and choose Run.
- Ribbon — select its Ribbon action. It starts with what's selected in the current tab. You can also run it from the Ribbon manager.
- Project folder — right-click the folder, open Tools, and choose the automation. Power Strip Folder panels for that folder list it too.
- Power Strip Automation command — run it from the command's sidecar, or drop files, folders, or text on the button to use them as input.
- Run Automation browser command — with Chrome in front and the extension connected, Power Strip captures the active page and runs the automation on it.
These runs happen on the desktop. The Chrome extension only supplies the page.
Scheduled and File-Watch Runs
Each placement can have its own schedules and file watches, set up the same way as for a Tool. A schedule runs with the placement's saved input. A file watch runs with the file that was added or modified. When the watch is for deleted files, the automation receives the deleted file's former path and name, not its contents.
ThinkForge must be running for these to fire, and unattended runs approve confirmation requests automatically. See Triggers.
How Paths Are Chosen
- Always (fan-out) lines all run, and each path gets the same result. They run one after another rather than at the same time.
- Conditional lines are checked in priority order. The path for the first rule that matches runs. If no rule matches, the Default route runs. Cards on paths that weren't taken are marked as skipped.
- A Merge card waits until every line into it has either run or been skipped, then passes on the results from the paths that ran.
Per-Item Runs
A Tool card set to For each item runs once for every item in its list, in order, then passes the collected results to the next card. If the path doesn't point to a list, the list is longer than the card's maximum, or a single item fails, the automation stops.
When Something Fails
The automation stops starting new steps when:
- A Tool it uses has been deleted.
- A Tool reports an error.
- A per-item run fails or its list isn't valid.
There's no automatic retry, no separate failure path, and no way to resume partway through. The Automations tab has no Stop button.
Seeing the Result
The Tools page and the Ribbon or folder runner show whether the automation finished or failed, along with its final result. Power Strip reports the result in its normal feedback, and each Tool the automation ran also appears in Power Strip Activity.
When one Output card ran, you get its result. When more than one ran, their results are collected under each Output card's name.
Check the files the automation produced as well as the success message, especially for work that matters.