Gpt Computer Assistant

🚀 Upsonic

Upsonic is a powerful tool for quickly building, deploying, and managing intelligent AI agents. It supports multiple models through the Model Context Protocol (MCP) and allows for easy integration of custom tools.

🚀 Quick Start

from upsonic import Agent, Task

# Create a simple task
task = Task("What is the capital of France?")

# Initialize the agent
agent = Agent("General Knowledge")

# Execute the task and print the result
agent.print_do(task)

✨ Features

• User-Friendly: Easy to get started.
• Modular Design: Flexible for expansion.
• Multilingual Support: Built - in support for multiple language models.
• Powerful Task Distribution Mechanism: Efficiently handle complex tasks.
• Structured Output: Define clear response formats via Pydantic.

💻 Usage Examples

Basic Usage

from upsonic import Agent, Task

# Create a simple task
task = Task("What is the capital of France?")

# Initialize the agent
agent = Agent("General Knowledge")

# Execute the task and print the result
agent.print_do(task)

Advanced Usage

1. Using MCP Tools - Fetch MCP Configuration

from upsonic import Agent, Task, ObjectResponse

class FetchMCP:
    command = "uvx"
    args = ["mcp-server-fetch"]
    
# Initialize the agent
web_agent = Agent(
    "Content Analyzer",
    model="openai/gpt-4o"  # Supports multiple models
)

# Create a task
task = Task(
    description="Fetch and analyze content from a URL, extract main information, title, and a brief summary.",
    context=["https://upsonic.ai/"],
    tools=[FetchMCP],
    response_format=WebContent
)

# Execute the task
result = web_agent.print_do(task)
print(result.title)  # Print the title
print(result.summary)  # Print the summary

2. Multi - Task Agent

from upsonic import Agent, Task, MultiAgent, ObjectResponse
from upsonic.tools import Search
import os

class CompanyResearch(ObjectResponse):
    industry: str
    product_focus: str
    company_values: list[str]
    recent_news: list[str]

class Mail(ObjectResponse):
    subject: str
    content: str

# Initialize agents
researcher = Agent("Company Researcher", company_url="https://redis.io/")
strategist = Agent("Outreach Strategist", company_url="https://redis.io/")

# Create tasks
company_task = Task(
    "Analyze key information on the company website",
    context=["https://upsonic.ai/"],
    tools=[Search],
    response_format=CompanyResearch
)

position_task = Task(
    "Analyze the requirements and background of a senior developer position",
    context=[company_task, "https://upsonic.ai/"]
)

message_task = Task(
    "Create a personalized outreach message based on the research results",
    context=[company_task, position_task],
    response_format=Mail
)

# Assign tasks to agents
results = MultiAgent.do([researcher, strategist], [company_task, position_task, message_task])

# Print the results
print(f"Company Industry: {company_task.response.industry}")
print(f"Product Focus: {company_task.response.product_focus}")
print(f"Company Values: {company_task.response.company_values}")
print(f"Recent News: {company_task.response.recent_news}")
print(f"Position Analysis Result: {position_task.response}")
print(f"Outreach Subject: {message_task.response.subject}")
print(f"Outreach Content: {message_task.response.content}")

3. Direct LLM Call

from upsonic import Agent, Task

# Create a task to directly query the model
task = Task("Explain the basic principles of quantum computing")

# Initialize the agent, specifying direct use of the LLM
agent = Agent("General Knowledge", mode="direct")

# Execute the task and print the result
agent.print_do(task)

4. Telemetry Setup

from upsonic import configure

# Configure telemetry (optional)
configure(metrics_enabled=True, logging_enabled=True)

📚 Documentation

Installation

pip install upsonic

Configuration

Refer to the official documentation for detailed settings.

Community Support

Join the discussion group.

📞 Contact

For more information, please visit the Upsonic official website or contact us via email: contact@upsonic.ai.