> Source: https://vitna.ai/docs/cli/aperture/triggers
> Raw Markdown: https://vitna.ai/docs/cli/aperture/triggers.md
> Corpus index: https://vitna.ai/llms.txt

# Trigger commands

Arming a watch on the record, and seeing what has fired.

```bash
npm run aperture:trigger:arm -- --workload <id> --kind price_move --threshold-pct 10
npm run aperture:trigger:disarm -- <id> --reason "..."
npm run aperture:triggers
npm run aperture:watch
```

## Arming

Refused where the kind names one sku and the workload names no incumbent. A
`price_move` watch with nothing to watch the price of is not a watch.

## Firing spends a trigger

A trigger fires once. It will not fire again until a decision arms a new one.

That is what stops one volatile cell producing the same notification every hour
until somebody mutes it, and a muted watch is a watch nobody reads. The cost is
that a second move during an unresolved review does not notify, which is the
right trade: the first one is already waiting for a person.

## Disarming takes a reason

Kept on the row. Calling a watch off is a decision about what to stop paying
attention to, and the register keeps it rather than the trigger quietly
disappearing.

## Listing

`aperture:triggers` prints what each workload watches and what has already
fired, with the event that fired it and the proposal it produced.

The same view is public at [`/api/v1/watches`](../../api/v1/engine/watches.md), and the
sentence describing a watch comes from the same function, so the CLI and the API
cannot describe one watch two ways.
