agenticmem 0.1.1

A Python client for the AgenticMem API

AgenticMem Python Client

A Python client library for interacting with the AgenticMem API. This client provides easy-to-use interfaces for managing user interactions and profiles.

Installation

pip install agenticmem

Quick Start

from agenticmem import AgenticMemClient
from agenticmem_commons.api_schema.service_schemas import InteractionRequest
from agenticmem_commons.api_schema.retriever_schema import (
    SearchInteractionRequest, 
    SearchUserProfileRequest,
    GetInteractionsRequest,
    GetUserProfilesRequest
)
from datetime import datetime

# Initialize the client
client = AgenticMemClient(api_key="your_api_key")

# Optional: Login with email/password
token = client.login(email="[e-mail]", password="password123")

# Publish a user interaction
interaction = InteractionRequest(
    timestamp=int(datetime.utcnow().timestamp()),
    text_interaction="User clicked on product X",
    user_action="click",
    user_action_description="Clicked on product details button"
)

response = client.publish_interaction(
    user_id="user123",
    request_id="req456",
    interaction_requests=[interaction]
)
print(f"Published interaction: {response.success} - {response.message}")

# Search user profiles
profiles_request = SearchUserProfileRequest(
    user_id="user123",
    search_query="recent interactions",
    top_k=5
)
profiles = client.search_profiles(profiles_request)
for profile in profiles.profiles:
    print(f"Profile {profile.profile_id}: {profile.profile_content}")

# Get user profiles directly
profiles_request = GetUserProfilesRequest(
    user_id="user123"
)
profiles = client.get_profiles(profiles_request)
for profile in profiles.profiles:
    print(f"Profile: {profile}")

# Search interactions
interactions_request = SearchInteractionRequest(
    user_id="user123",
    start_time=int(datetime(2024, 1, 1).timestamp()),
    end_time=int(datetime.utcnow().timestamp())
)
interactions = client.search_interactions(interactions_request)
for interaction in interactions.interactions:
    print(f"Interaction {interaction.interaction_id}: {interaction.text_interaction}")

# Get interactions directly
interactions_request = GetInteractionsRequest(
    user_id="user123"
)
interactions = client.get_interactions(interactions_request)
for interaction in interactions.interactions:
    print(f"Interaction: {interaction}")

# Get profile change log
change_log = client.get_profile_change_log()
print(f"Profile changes: {change_log}")

Features

• Authentication
  • API key authentication
  • Email/password login
• User interaction management
  • Publish user interactions
  • Delete specific interactions
  • Search interactions with time range and filters
  • Get direct list of interactions
• User profile management
  • Search user profiles with customizable queries
  • Get direct list of user profiles
  • Delete specific profiles or profiles matching a search query
  • View profile change log history

API Response Types

All API methods return strongly-typed responses:

• login() returns Token
• publish_interaction() returns PublishUserInteractionResponse
• search_interactions() returns SearchInteractionResponse
• get_interactions() returns GetInteractionsResponse
• search_profiles() returns SearchUserProfileResponse
• get_profiles() returns GetUserProfilesResponse
• delete_profile() returns DeleteUserProfileResponse
• delete_interaction() returns DeleteUserInteractionResponse
• get_profile_change_log() returns ProfileChangeLogResponse

Documentation

For detailed documentation, please visit [docs link].

License

MIT License