Event tags are a powerful feature for automatically categorizing user events with semantic attributes. You can use them to enrich event data and track user behavior over time across various dimensions, such as:
  • Emotion (happy, frustrated)
  • Life Goals (buying_a_house, learning_a_skill)
  • Relationships (new_friend, family_mention)

​
Configuring Event Tags

By default, no event tags are recorded. You must define the tags you want to track in your config.yaml file:
config.yaml
event_tags:
  - name: "emotion"
    description: "Records the user's current emotional state."
  - name: "romance"
    description: "Tracks any mention of romantic relationships or feelings."
Once configured, Memobase will automatically analyze interactions and apply these tags to events when relevant. The description field is crucial for helping the AI accurately understand when to apply each tag.

​
Retrieving Event Tags

Event tags are returned as part of the event object when you retrieve a user’s event history. 
from memobase import MemoBaseClient

client = MemoBaseClient(project_url='YOUR_PROJECT_URL', api_key='YOUR_API_KEY')
user = client.get_user('some_user_id')

events = user.event()
for event in events:
    print(event.event_data.event_tags)

​
Searching Events by Tag

You can search for events that have specific tags applied using the search_event_by_tags method. 
from memobase import MemoBaseClient

client = MemoBaseClient(project_url='YOUR_PROJECT_URL', api_key='YOUR_API_KEY')
user = client.get_user('some_user_id')

# Find events with specific tags (AND condition)
events = user.search_event_by_tags(tags=["emotion", "romance"])
print(events)
For more details, see the API Reference.