How It Works

Every poll cycle follows this decision flow:

1. Load configuration from --env-file.
2. Resolve the PagerDuty user from PD_USER_ID or /users/me.
3. Call GET /oncalls for that user at the current time.
4. Stop immediately if the user is not currently on call.
5. Call GET /incidents for open incidents assigned to that user with statuses[]=triggered and statuses[]=acknowledged.
6. Display the open incident list.
7. Stop if there are no matching triggered incidents.
8. In dry-run mode, print what would be acknowledged.
9. In live mode, call PUT /incidents/{id} with status: acknowledged.

The CLI only acknowledges incidents when:

• the configured user is currently on call
• the incident is triggered
• the incident is assigned to that user

The CLI does not acknowledge incidents assigned to someone else. It does not acknowledge already acknowledged incidents. It does not resolve incidents, reassign incidents, edit titles, or add notes.

Open vs Triggered

For display, open incidents means incidents with PagerDuty status triggered or acknowledged.

For action, only triggered incidents are eligible for acknowledgement. Already acknowledged incidents are shown for operator awareness but are not updated.

Dry-Run vs Live Mode

Dry-run mode is the default. These commands do not change PagerDuty:

pd-auto-ack run --env-file ~/.pd-auto-ack.env --once
pd-auto-ack run --env-file ~/.pd-auto-ack.env --watch --interval 30

Live mode requires --apply:

pd-auto-ack run --env-file ~/.pd-auto-ack.env --watch --interval 30 --apply

You can also set PD_APPLY=true in the env file, but leaving PD_APPLY=false and passing --apply explicitly is safer.

Terminal Dashboard

pd-auto-ack tui uses the same decision flow as run, but displays it in an interactive terminal dashboard.

The dashboard refreshes on the configured interval, shows open incidents and selected incident details, and starts in dry-run mode unless live mode is explicitly enabled with --apply or PD_APPLY=true.