Home Assistant integration for the PreCom fire department alerting service. Monitors incoming P2000 alarms, shows staffing levels per function at your fire station, and lets you report your availability (inside or outside your response region) directly from Home Assistant.

• Home Assistant 2024.1.0 or newer
• A PreCom account (username + password)
• HACS installed

1. Open HACS → Integrations.
2. Click the three-dot menu → Custom repositories.
3. Add this repository URL and select category Integration.
4. Search for PreCom and click Download.
5. Restart Home Assistant.

1. Copy the custom_components/precom folder into your <config>/custom_components/ directory.
2. Restart Home Assistant.

1. Go to Settings → Devices & Services → Add Integration.
2. Search for PreCom.
3. Enter your PreCom username and password.
4. Optionally set a scan interval (range: 10–3600 s). Leave blank to disable automatic polling — use precom.update_alarm to refresh on demand instead.

Credentials are validated against the PreCom API before the entry is saved.

One sensor is created per group (e.g. sensor.precom_velp_frb_en_vk_alarm). The sensor shows the latest alarm for that specific group from the past 30 days. The state is the alarm message text or none when no recent alarm is found.

Note: Group alarm sensors use portal scraping with a 30-day lookback window. This allows viewing historical alarms beyond the API's retention period. Data is updated every scan interval.

One binary sensor is created per function (e.g. binary_sensor.precom_staffing_chauffeur_ts). The sensor is on (problem) when any 15-minute slot in the next 24 hours has fewer available people than required.

Mark yourself as unavailable (outside region) for a number of hours. Targets the binary_sensor.precom_availability entity.

Cancel the unavailable status and mark yourself as available again. Targets the binary_sensor.precom_availability entity. No fields required.

Force an immediate refresh of the latest alarm data from PreCom, without waiting for the next scheduled poll interval. Targets the sensor.precom_last_alarm entity. No fields required.

Look up portal report details for a specific alarm message text and return portal response data, staffing requirements, and proposed functions.

automation:
  - alias: "Notify on new PreCom alarm"
    trigger:
      - platform: state
        entity_id: sensor.precom_last_alarm
    condition:
      - condition: template
        value_template: "{{ trigger.to_state.state != 'none' }}"
    action:
      - service: notify.mobile_app_your_phone
        data:
          title: "PreCom Alarm"
          message: "{{ states('sensor.precom_last_alarm') }}"

• Invalid credentials on setup — verify your username and password at portal.pre-com.nl.
• Sensor stuck on none — check that your account has access to alarm messages and review the Home Assistant logs for API errors.
• No staffing sensors appearing — the sensors are created dynamically after the first successful poll. Reload the integration if they do not appear after a minute.
• Rate limiting / connectivity — increase the scan interval in the integration options.
• "Authentication failed" after working correctly — Home Assistant will automatically prompt you to re-enter your password via a re-authentication notification. Go to Settings → Integrations → PreCom and follow the notification.
• Reconfiguring after an email/password change — click the three-dot menu on the integration card → Reconfigure, enter new credentials.

The integration uses a polling model (iot_class: cloud_polling). Every scan_interval seconds (default 60 s, configurable 10–3600 s) the coordinator fetches alarm data, user info, and staffing data from the PreCom API. For staffing, both today's and tomorrow's data are fetched per group so that the next 24 hours can be evaluated at 15-minute granularity. The JWT token is cached in memory; if it is rejected (HTTP 401) the coordinator re-authenticates once and retries before marking the sensor unavailable.

No webhooks or push mechanisms are used.

• Alarm notifications — trigger a mobile notification, flash a light, or start a siren whenever sensor.precom_last_alarm changes to a non-none value.
• Group-specific alarms — monitor individual groups with sensor.precom_<group>_alarm to track alarms per station or unit, with 30-day history.
• Automatic availability reporting — use a device tracker or input boolean to automatically call precom.set_unavailable when you leave home.
• Dashboard card — display the latest alarm text and staffing levels on a Lovelace dashboard.
• Understaffing alerts — use the staffing binary sensors to send a notification when your fire station is short on a specific function in the coming hours.
• Response tracking — combined with precom.set_available / precom.set_unavailable, build automations that keep PreCom in sync with your physical location.

• Polling only — no push support; minimum latency equals the configured scan interval.
• No P2000 raw messages — alarm data is what PreCom exposes via its own API; raw P2000 data is not available.
• Single device per entry — each config entry covers one PreCom account. Multiple accounts require multiple entries.
• Limited alarm history — sensor.precom_last_alarm shows only the current active alarm. Group sensors (sensor.precom_<group>_alarm) show the latest alarm from the past 30 days per group via portal scraping.
• Dynamic sensors — group sensors and staffing sensors are created dynamically after the first successful poll. Removing a group or function from PreCom does not automatically remove the corresponding entity from Home Assistant.
• Portal scraping overhead — group alarm sensors scrape the portal per group per update cycle. For accounts with many groups, consider increasing the scan interval to reduce load.

1. Go to Settings → Devices & Services.
2. Find the PreCom integration and click the three-dot menu → Delete.
3. Confirm deletion. Home Assistant will unload the integration and remove all associated entities and devices.
4. No files are left behind after removal.

Issues and pull requests are welcome. Please open an issue before submitting large changes.