https://github.com/upsonic/upsonic
The most reliable AI agent framework that supports MCP.
https://github.com/upsonic/upsonic
agent agent-framework claude computer-use llms mcp model-context-protocol openai rag reliability
Last synced: about 1 month ago
JSON representation
The most reliable AI agent framework that supports MCP.
- Host: GitHub
- URL: https://github.com/upsonic/upsonic
- Owner: Upsonic
- License: mit
- Created: 2024-05-26T16:23:38.000Z (over 1 year ago)
- Default Branch: master
- Last Pushed: 2025-05-01T17:11:21.000Z (7 months ago)
- Last Synced: 2025-05-01T17:49:53.512Z (7 months ago)
- Topics: agent, agent-framework, claude, computer-use, llms, mcp, model-context-protocol, openai, rag, reliability
- Language: Python
- Homepage: https://docs.upsonic.ai
- Size: 4.01 MB
- Stars: 7,398
- Watchers: 56
- Forks: 689
- Open Issues: 44
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
- acu - Upsonic
- awesome-agents - Upsonic
- awesome-ChatGPT-repositories - Upsonic - We do NOT and WILL not have any Crypto Projects, they are a complete SCAM | Task oriented AI agent framework for digital workers and vertical AI agents (Chatbots)
- awesome-llm-agents - Upsonic - Reliable agent framework that (Frameworks)
- Awesome-AI-Agents - Upsonic - Upsonic is a reliable agent framework supporting MCP, offering trusted agent workflows with verification layers.  (Frameworks / Advanced Components)
- awesome-ai-agents - Upsonic - oriented | (🌟 Core Frameworks)
- awesome_ai_agents - Upsonic - Reliable agent framework that supports MCP. [github](https://github.com/upsonic/upsonic) (Building / Frameworks)
README
# Introduction
Upsonic is an AI agent development framework used by fintech leaders and tested at their scale against attacks and reasoning puzzles.
```bash
pip install upsonic
```
```python
from upsonic import Task, Agent
task = Task("Who developed you?")
agent = Agent(name="Coder")
agent.print_do(task)
```
# Guides | 7 Step
See our guides to jumpstart your AI agent within minutes. We design them to onboard the new users to the framework.
1. [Create an Agent](https://docs.upsonic.ai/guides/1-create-a-task)
2. [Create a Task](https://docs.upsonic.ai/guides/2-create-an-agent)
3. [Add a Safety Engine](https://docs.upsonic.ai/guides/3-add-a-safety-engine)
4. [Add a Tool](https://docs.upsonic.ai/guides/4-add-a-tool)
5. [Add an MCP](https://docs.upsonic.ai/guides/5-add-an-mcp)
6. [Integrate a Memory](https://docs.upsonic.ai/guides/6-integrate-a-memory)
7. [Creating a Team of Agents](https://docs.upsonic.ai/guides/7-creating-a-team-of-agents)
# Why Upsonic?
Upsonic provides a feature set to build safety-first, high-performance AI Agents. It helps you go to production without spending hours on research and boilerplate. These are the main parts:
- **Safety First**: Upsonic provides its own **Safety Engine** that manages User and Agent messages and checks their status for your policies. You can customize it by designing new **rule** and **action** sets.
- **Direct LLM Calls**: In Upsonic we support the same interface for your whole AI operations. You don't need to go with another framework to complete your **small jobs**.
- **Structured Outputs**: Upsonic sets agent outputs to make them **Python objects**. So you can integrate your application without struggling with **LLM outputs**.
- **Built-in RAG and Memory**: In Upsonic you can create world class . We support the Agentic RAG, Memory Logics and providers of them.
- **Customizable Memory Logics**: You are able to create **memories** that focus on **user**, **event** and **chat**. Also you are free to use **Local** and **Cloud databases**.
- **Agent Teams**: Upsonic provides the most **reliable** agent team architecture with memory, context management and leader agent.
- **FastAPI Compatible Agents**: You can turn your agents into production-ready APIs
- **Tracking the Executions**: You can use Upsonic AgentOS to get the execution history, monthly costs andresponse times of your agents.
- **Deploy at scale**: Upsonic agents work in the greatest and fastest-growing fintech companies and scaling is available on Upsonic AgentOS.
# 📙 Documentation
You can access our documentation at [docs.upsonic.ai](https://docs.upsonic.ai/) All concepts and examples are available there.
## 📊 Telemetry & Privacy
Upsonic uses **anonymous telemetry** to help us understand how the framework is being used and improve our development focus. We are committed to transparency and user privacy.
All telemetry is **anonymous** - we only track a randomly generated system ID to distinguish unique installations.
### Why Collect Telemetry?
Telemetry helps us:
- 🎯 Focus development on frequently-used features
- 🐛 Identify and fix common errors and edge cases
- 📈 Understand performance characteristics at scale
- 🔧 Improve framework reliability
### How to Disable Telemetry
You can **completely disable** telemetry in multiple ways:
**Option 1: Environment Variable (Recommended)**
```bash
export UPSONIC_TELEMETRY=false
```
**Option 2: In Python Code**
```python
import os
os.environ["UPSONIC_TELEMETRY"] = "false"
from upsonic import Agent # Import after setting env var
```
**Option 3: .env File**
```bash
# .env
UPSONIC_TELEMETRY=false
```
Once disabled, **no data** will be sent to our telemetry service.