# Upsonic

> MCP Client

Upsonic is a reliability-focused framework for building trusted agent workflows, offering features like verification layers and a Model Context Protocol for leveraging diverse tools. It enables secure agent deployment in real-world applications.

## Overview

- **Category:** AI
- **Language:** Python
- **Stars:** 7923
- **Forks:** 670
- **Owner:** Upsonic
- **GitHub:** https://github.com/Upsonic/Upsonic
- **Homepage:** https://upsonic.ai
- **Created:** 2024-05-26T16:23:38+00:00
- **Updated:** 2025-03-28T19:36:30+00:00
- **Source:** https://model-context-protocol.com/clients/reliable-ai-agent-framework-mcp

## Setup

## Setup

### Prerequisites

- Python 3.10 or higher
- Access to OpenAI or Anthropic API keys (Azure and Bedrock Supported)

## Installation

```bash
pip install upsonic
```

Set your OPENAI_API_KEY

```console
export OPENAI_API_KEY=sk-***
```

## Tools

## Available Tools

1.  **Reliability Layer** (Offers easy-to-activate reliability layers without disrupting functionality.)
2.  **Model Context Protocol (MCP)** (Allows leveraging tools with various functionalities developed officially and by third parties.)
3.  **Integrated Browser Use and Computer Use** (Directly use and deploy agents that works on non-API systems.)
4.  **Secure Runtime** (Isolated environment to run agents.)
5.  **Verifier Agent** (Validates outputs, tasks, and formats - detecting inconsistencies, numerical errors, and hallucinations.)
6.  **Editor Agent** (Works with verifier feedback to revise and refine outputs until they meet quality standards.)
7.  **Rounds** (Implements iterative quality improvement through scored verification cycles.)
8.  **Loops** (Ensures accuracy through controlled feedback loops at critical reliability checkpoints.)
9.  **Production-Ready Scalability** (Deploy seamlessly on AWS, GCP, or locally using Docker.)
10. **Task-Centric Design** (Focus on practical task execution with options for basic, advanced, and complex automation.)
11. **MCP Server Support** (Utilize multi-client processing for high-performance tasks.)
12. **Tool-Calling Server** (Exception-secure tool management with robust server API interactions.)
13. **Computer Use Integration** (Execute human-like tasks using Anthropic’s ‘Computer Use’ capabilities.)
14. **Easily adding tools** (Add custom tools and MCP tools with a single line of code.)
15. **Automated task distribution mechanism** (Matches tasks based on the relationship between agent and task, ensuring collaborative problem-solving.)
16. **Direct LLM Calls** (Make calls to model providers without any abstraction level and organize structured outputs.)