86 lines
2.6 KiB
Plaintext
86 lines
2.6 KiB
Plaintext
---
|
|
title: MultiOn
|
|
---
|
|
|
|
Build personal browser agent remembers user preferences and automates web tasks. It integrates Mem0 for memory management with MultiOn for executing browser actions, enabling personalized and efficient web interactions.
|
|
|
|
## Overview
|
|
|
|
In this example, we will create a Browser based AI Agent that searches [arxiv.org](https://arxiv.org) for research papers relevant to user's research interests.
|
|
|
|
## Setup and Configuration
|
|
|
|
Install necessary libraries:
|
|
|
|
```bash
|
|
pip install mem0ai multion
|
|
```
|
|
|
|
First, we'll import the necessary libraries and set up our configurations.
|
|
|
|
```python
|
|
import os
|
|
from mem0 import Memory
|
|
from multion.client import MultiOn
|
|
|
|
# Configuration
|
|
OPENAI_API_KEY = 'sk-xxx' # Replace with your actual OpenAI API key
|
|
MULTION_API_KEY = 'your-multion-key' # Replace with your actual MultiOn API key
|
|
USER_ID = "deshraj"
|
|
|
|
# Set up OpenAI API key
|
|
os.environ['OPENAI_API_KEY'] = OPENAI_API_KEY
|
|
|
|
# Initialize Mem0 and MultiOn
|
|
memory = Memory()
|
|
multion = MultiOn(api_key=MULTION_API_KEY)
|
|
```
|
|
|
|
## Add memories to Mem0
|
|
|
|
Next, we'll define our user data and add it to Mem0.
|
|
|
|
```python
|
|
# Define user data
|
|
USER_DATA = """
|
|
About me
|
|
- I'm Deshraj Yadav, Co-founder and CTO at Mem0, interested in AI and ML Infrastructure.
|
|
- Previously, I was a Senior Autopilot Engineer at Tesla, leading the AI Platform for Autopilot.
|
|
- I built EvalAI at Georgia Tech, an open-source platform for evaluating ML algorithms.
|
|
- Outside of work, I enjoy playing cricket in two leagues in the San Francisco.
|
|
"""
|
|
|
|
# Add user data to memory
|
|
memory.add(USER_DATA, user_id=USER_ID)
|
|
print("User data added to memory.")
|
|
```
|
|
|
|
## Retrieving Relevant Memories
|
|
|
|
Now, we'll define our search command and retrieve relevant memories from Mem0.
|
|
|
|
```python
|
|
# Define search command and retrieve relevant memories
|
|
command = "Find papers on arxiv that I should read based on my interests."
|
|
|
|
relevant_memories = memory.search(command, user_id=USER_ID, limit=3)
|
|
relevant_memories_text = '\n'.join(mem['text'] for mem in relevant_memories)
|
|
print(f"Relevant memories:")
|
|
print(relevant_memories_text)
|
|
```
|
|
|
|
## Browsing arXiv
|
|
|
|
Finally, we'll use MultiOn to browse arXiv based on our command and relevant memories.
|
|
|
|
```python
|
|
# Create prompt and browse arXiv
|
|
prompt = f"{command}\n My past memories: {relevant_memories_text}"
|
|
browse_result = multion.browse(cmd=prompt, url="https://arxiv.org/")
|
|
print(browse_result)
|
|
```
|
|
|
|
## Conclusion
|
|
|
|
By integrating Mem0 with MultiOn, you've created a personalized browser agent that remembers user preferences and automates web tasks. For more details and advanced usage, refer to the full [cookbook here](https://github.com/mem0ai/mem0/blob/main/cookbooks/mem0-multion.ipynb).
|