> ## Documentation Index
> Fetch the complete documentation index at: https://docs.openbunny.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Reminders

> Automated task reminders with decision and schedule engines.

OpenBunny has two reminder systems that work together to keep tasks on track.

## Decision engine

The decision engine periodically reviews all open tasks using the LLM. It reads a configurable policy file (`data/reminder-policy.md`) and decides which tasks need attention, escalation, or follow-up.

This runs on a timer and produces reminder actions that are logged for auditing.

## Schedule engine

The schedule engine handles per-task reminders that users configure explicitly. Each task can have one or more reminder schedules.

### Schedule types

| Type     | Description                                | Example                        |
| -------- | ------------------------------------------ | ------------------------------ |
| One-time | Fires once at a specific date/time         | "Remind me on March 15 at 9am" |
| Interval | Repeats every N days, weeks, or months     | "Every 3 days"                 |
| Weekly   | Fires on a specific day and time each week | "Every Monday at 10am"         |
| Monthly  | Fires on a specific date each month        | "1st of every month at 9am"    |

### Managing schedules

Schedules are managed through the REST API or the client UI:

```bash theme={null}
# Create a reminder
curl -X POST http://localhost:3100/tasks/123/schedules \
  -H "Content-Type: application/json" \
  -d '{"type": "interval", "interval_days": 3}'

# List reminders for a task
curl http://localhost:3100/tasks/123/schedules

# Delete a reminder
curl -X DELETE http://localhost:3100/tasks/123/schedules/456
```

## Reminder policy

The reminder policy file (`data/reminder-policy.md`) is a markdown document that instructs the LLM on how to evaluate tasks for reminders. You can customize it to match your team's workflow — for example, escalating high-priority tasks that have been idle for more than 24 hours.