Worker setup

Run it where you have control.

Use a reviewed checkout and your own model access. The browser prepares files; the worker executes the tasks you authorize.

Install the local tools

Use a Linux, macOS or WSL worker with Python and Git. Keep the checkout because the queue is a repository tool.

git clone https://github.com/alisonjieli-png/loop-engine.git
cd loop-engine
python3 -m venv .venv
.venv/bin/python -m pip install -e .

Configure your model access
Choose an endpoint you can reach, including your own machine, network service or Ollama Cloud. A local worker does not require a local model download.
```
.venv/bin/python -m loop_engine configure
.venv/bin/python tools/overnight_queue.py --help
```
Use the provider and credential guide. Keep keys on the worker, not in task files or this website.
Write one short task
State the objective, permitted inputs and changes, destination files, and checks for success. Do not include work whose prerequisite result still needs approval.
A useful task file includes
- Objective and expected output
- Allowed files and effects
- Time and provider limits
- Checks that determine acceptance
Prepare and inspect the queue
Enter the absolute task-file path in the run desk. Download the manifest, save it in your checkout, and inspect the generated command before running it.
Open the run desk →
Review the result before leaving more work
Check the worker’s status and output files. A finished process is not proof that the task succeeded. Preserve the original run folder for interrupted or uncertain work.