# Upsonic/Upsonic

Build autonomous AI agents in Python.

Repository: https://github.com/Upsonic/Upsonic
Canonical: https://ross.abutalabs.com/products/upsonic
Homepage: https://docs.upsonic.ai
Language: Python
License: MIT
License Family: permissive
Topics: openai, computer-use, claude, mcp, agent-framework, agent, llms, reliability, model-context-protocol, rag, ucp, universal-commerce-protocol, autonomous-agent, autonomous-agents, openclaw
Last push: 2026-06-18T14:55:50+00:00

## Health v2 (maintenance only)
Score: 77/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 88, release rhythm 72, longevity 59
- inputs: {"age_days": 829, "days_push": 76, "days_rel": 106, "gap_med": 0, "n_releases_24m": 222}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7948, forks 745 (observed 2026-08-28T04:10:10.143599+00:00)

## What it is
Upsonic is a Python framework for building autonomous and traditional AI agents, with support for tools, memory, knowledge bases, RAG, multi-agent teams, and production deployment. It provides a unified API over 30+ LLM providers and includes prebuilt autonomous agents that can plan and execute shell and filesystem operations in a sandboxed workspace.

## Use cases
- build autonomous AI agents in python
- create a multi-agent team with tools and memory
- add RAG knowledge base to an llm agent
- run an agent that analyzes server logs and executes shell commands
- connect an agent to openai, anthropic, or ollama models
- build a stock analyst agent with yfinance tools
- deploy production ai agents with reliability layer

## When to choose
- you want a Python framework for autonomous, multi-step agents with sandboxed execution
- you need provider-agnostic LLM support across 30+ providers including local models
- you want built-in memory, knowledge bases, RAG, and multi-agent teams in one API
- you want prebuilt community agents to run immediately

## When to avoid
- you need a simple single LLM call without agent orchestration
- you require a non-Python language
- you need a fully mature, long-proven framework for critical production systems
- you want a no-code or UI-first agent builder

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, rag, llm-inference, mcp, chatbot, sdk
- domain: artificial-intelligence, large-language-models, machine-learning, developer-tools
- platform: python, cross-platform
- tags: autonomous-agents, multi-agent, computer-use, llm-providers, memory, knowledge-base, openai, anthropic, ollama, universal-commerce-protocol, ai-agents, retrieval-augmented-generation

## Member repositories
- Upsonic/Upsonic (main) score 77

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:10.143599+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:33:12.666550+00:00, confidence not recorded.
  - readme: https://github.com/Upsonic/Upsonic (fetched 2026-08-28T04:10:10.143599+00:00, sha 4cf183b80239)
  - homepage: https://docs.upsonic.ai (fetched 2026-08-29T08:29:52.030132+00:00, sha 05bca5143d35)
  - site_page: https://docs.upsonic.ai/integrations/overview (fetched 2026-08-29T08:29:52.039560+00:00, sha 80458c546716)
  - site_page: https://docs.upsonic.ai/changelog (fetched 2026-08-29T08:29:52.041637+00:00, sha de341d50bfe1)
  - site_page: https://docs.upsonic.ai/get-started/quickstart (fetched 2026-08-29T08:29:52.044459+00:00, sha ed53e9e965b0)
  - site_page: https://docs.upsonic.ai/get-started/installation (fetched 2026-08-29T08:29:52.046334+00:00, sha 3ff66ce4c5ca)
- Data as of 2026-08-30T08:39:29.467469+00:00.