Fork of valentinfrlch/ha-llmvision blueprints/event_summary.yaml to support other notification platforms

ha_blueprint-llm_vision_event_summary.yaml

blueprint:
  name: AI Event Summary (LLM Vision)
  author: valentinfrlch
  description: >
    AI-powered security event summaries for frigate or camera entities. 
    Sends a notification with a preview to your phone that is updated dynamically when the AI summary is available.
  domain: automation
  input:
    mode:
      name: Mode
      description: Select the mode to use
      selector:
        select:
          options:
            - 'Frigate'
            - 'Camera'
    important:
      name: Important
      description: "Use AI to notify only on important events. Use with caution: AI can make mistakes."
      default: false
      selector:
        boolean:
    notify_device:
      name: Notify Device
      description: The device to send the notification to
      selector:
        device:
          integration: mobile_app
    frigate_url:
      name: Frigate URL
      description: (Frigate) Frigate's base url to fetch preview
      default: "http://localhost:5000"
      selector:
        text:
          multiline: false
    camera_entities:
      name: Camera Entities
      description: (Camera) List of camera entities to monitor
      default: []
      selector:
        entity:
          multiple: true
          filter:
            domain: camera
    motion_sensors:
      name: Motion Sensor
      description: (Camera) Set if your cameras don't change state. Use the same order used for camera entities.
      default: []
      selector:
        entity:
          multiple: true
          filter:
            domain: binary_sensor
    trigger_state:
      name: Trigger State
      description: (Camera) Trigger the automation when your cameras change to this state.
      default: 'recording'
      selector:
        text:
          multiline: false
    tap_navigate:
      name: Tap Navigate
      description: Path to navigate to when tapping notification (e.g. /lovelace/cameras)
      default: "/lovelace/0"
      selector:
        text:
          multiline: false
    cooldown:
      name: Cooldown
      description: Time in minutes to wait before running again.
      default: 10
      selector:
        number:
          min: 0
          max: 60
    provider:
      name: Provider
      description: Configuration to use for the video_analyzer service. See docs for additional information.
      selector:
        config_entry:
          integration: llmvision
    max_frames:
      name: Max Frames
      description: How many frames to analyze. Picks frames with the most movement.
      default: 3
      selector:
        number:
          min: 1
          max: 60
    duration:
      name: Duration
      description: (Camera) How long to record for (in seconds)
      default: 5
      selector:
        number:
          min: 1
          max: 60
    model:
      name: Model
      description: Model to use for the video_analyzer service
      default: "gpt-4o-mini"
      selector:
        text:
          multiline: false
    message:
      name: Prompt
      description: Model prompt for the video_analyzer service
      default: "Summarize briefly what's happening in the camera feed (one sentence max). Don't describe the scene! If there is a person, describe what they're doing. If nothing is happening, say so."
      selector:
        text:
          multiline: true
    target_width:
      name: Target Width
      description: Width in pixels to downscale (uses less tokens)
      default: 1280
      selector:
        number:
          min: 512
          max: 3840
    detail:
      name: Detail
      description: Detail parameter (OpenAI only)
      default: 'low'
      selector:
        select:
          options:
            - 'high'
            - 'low'
    max_tokens:
      name: Maximum Tokens
      description: Maximum number of tokens to generate
      default: 20
      selector:
        number:
          min: 1
          max: 100
    temperature:
      name: Temperature
      description: Randomness. Lower is more accurate, higher is more creative
      default: 0.1
      selector:
        number:
          min: 0.1
          max: 1.0
          step: 0.1

variables:
  important: !input important
  base_url: !input frigate_url
  cooldown: !input cooldown
  mode: !input mode
  camera_entities_list: !input camera_entities
  motion_sensors_list: !input motion_sensors
  camera_entity: >
    {% if mode == 'Camera' %}
      {% if motion_sensors_list %}
        {% set index = motion_sensors_list.index(trigger.entity_id) %}
        {{ camera_entities_list[index] }}
      {% else %}
        {{ trigger.entity_id }}
      {% endif %}
    {% else %}
      {{ trigger.payload_json['after']['camera'] }}
    {% endif %}
  tag: >
    {% if mode == 'Frigate' %}
      {{ trigger.payload_json['after']['camera'] + int(as_timestamp(now()))|string }}
    {% else %}
      {{ camera_entity + int(as_timestamp(now()))|string }}
    {% endif %}
  group: >
    {% if mode == 'Frigate' %}
      {{ trigger.payload_json['after']['camera'] }}
    {% else %}
      {{ camera_entity }}
    {% endif %}
  label: >
    {% if mode == 'Frigate' %}
      {{ trigger.payload_json['after']['label']|capitalize }} seen
    {% else %}
      Motion detected
    {% endif %}
  camera: >
    {% if mode == 'Frigate' %}
      {{ trigger.payload_json['after']['camera'].replace('_', ' ')|capitalize }}
    {% else %}
      {{ camera_entity.replace("camera.", "").replace("_", " ")|capitalize }}
    {% endif %}
  video: >
    {% if mode == 'Frigate' %}
      {{ base_url }}/api/events/{{ trigger.payload_json['after']['id'] }}/clip.mp4
    {% else %}
      ''
    {% endif %}
  image: >
    {% if mode == 'Frigate' %}
      ''
    {% else %}
      {{ '/api/camera_proxy/' + camera_entity }}
    {% endif %}
  importance_prompt: >
    Your job is to decide whether a camera recording is important (so the user gets a notification) or not. 
    Reply with 'yes' if the user should be notified or 'no' otherwise. 
    Reply with these replies exactly.

trigger:
  - platform: mqtt
    topic: "frigate/events"
    id: frigate_trigger
  - platform: 'state'
    entity_id: !input camera_entities
    to: !input trigger_state
    id: 'camera_trigger'
  - platform: 'state'
    entity_id: !input motion_sensors
    to: 'on'
    id: 'motion_sensor_trigger'

condition:
  - condition: template
    value_template: >
      {% if mode == 'Frigate' %}
        {{ trigger.payload_json["type"] == "end" and (state_attr(this.entity_id, 'last_triggered') is none or (now() - state_attr(this.entity_id, 'last_triggered')).total_seconds() / 60 > cooldown) }}
      {% else %}
        {{ state_attr(this.entity_id, 'last_triggered') is none or (now() - state_attr(this.entity_id, 'last_triggered')).total_seconds() / 60 > cooldown }}
      {% endif %}

action:
  - choose:
      - conditions:
          - condition: template
            value_template: "{{ important }}"
        sequence:
          - alias: "Decide Important"
            choose:
              - conditions:
                  - condition: template
                    value_template: "{{ mode == 'Frigate'}}"
                sequence:
                  - service: llmvision.image_analyzer
                    data:
                      image_entity: "{{ ['camera.' + trigger.payload_json['after']['camera']|lower] }}"
                      provider: !input provider
                      model: !input model
                      message: "{{importance_prompt}}"
                      include_filename: true
                      target_width: 1280
                      detail: low
                      max_tokens: 1
                      temperature: 0.1
                    response_variable: importance
              - conditions:
                  - condition: template
                    value_template: "{{ mode == 'Camera' }}"
                sequence:
                  - service: llmvision.image_analyzer
                    data:
                      image_entity: "{{[camera_entity]}}"
                      provider: !input provider
                      model: !input model
                      message: "{{importance_prompt}}"
                      include_filename: true
                      target_width: 1280
                      detail: low
                      max_tokens: 1
                      temperature: 0.1
                    response_variable: importance
    
  # Cancel automation if event not deemed important
  - choose:
      - conditions:
          - condition: template
            value_template: "{{ importance is defined and importance.response_text == 'no' }}"
        sequence:
          - stop: "Event is not important"

  - choose:
      - conditions:
          - condition: template
            value_template: "{{ base_url is not none and base_url != '' }}"
        sequence:
          - alias: "Send instant notification without summary"
            domain: mobile_app
            type: notify
            device_id: !input notify_device
            title: "{{ label }}"
            message: "{{camera}}"
            data:
              video: "{{video}}"
              image: "{{image}}"
              entity_id: "{{camera_entity if mode=='Camera' else ''}}"
              url: !input tap_navigate #iOS
              clickAction: !input tap_navigate #Android
              tag: "{{tag}}"
              group: "{{group}}"

  - alias: "Analyze event"
    choose:
      - conditions:
          - condition: template
            value_template: "{{ mode == 'Frigate' }}"
        sequence:
          - service: llmvision.video_analyzer
            data:
              event_id: "{{ trigger.payload_json['after']['id'] }}"
              provider: !input provider
              model: !input model
              message: !input message
              include_filename: true
              max_frames: !input max_frames
              target_width: !input target_width
              detail: !input detail
              max_tokens: !input max_tokens
              temperature: !input temperature
            response_variable: response
      - conditions:
          - condition: template
            value_template: "{{ mode == 'Camera' }}"
        sequence:
          - service: llmvision.stream_analyzer
            data:
              image_entity: "{{[camera_entity]}}"
              duration: !input duration
              provider: !input provider
              model: !input model
              message: !input message
              include_filename: true
              max_frames: !input max_frames
              target_width: !input target_width
              detail: !input detail
              max_tokens: !input max_tokens
              temperature: !input temperature
            response_variable: response

  - alias: "Send notification with summary"
    domain: mobile_app
    type: notify
    device_id: !input notify_device
    title: "{{label}}"
    message: "{{ response.response_text }}"
    data:
      tag: "{{tag}}"
      group: "{{group}}"
      image: "{{image}}"
      video: "{{video}}"
      entity_id: "{{camera_entity if mode=='Camera' else ''}}"
      url: !input tap_navigate #iOS
      clickAction: !input tap_navigate #Android
      push:
        interruption-level: passive