M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
U

Upsonic

Community
Framework / SDK
Build autonomous AI agents in Python

This is a framework for building MCP servers or agents, not a server you connect a client to.

How to install ↓ Compare GitHub source repository ↗
Category
Dev Tools #394 of 438
Stars
★ 8.0k Very popular
Runtime
Python
Credentials
No credential needed
License
MIT
Last commit
34FMRS · D

The supplied material identifies Upsonic as an MIT-licensed Python AI-agent framework, not as a clearly defined MCP server. The README only states that MCP tools can be integrated; it provides no evidence for a server manifest, tools, transport, authentication, or client configuration.

Strongest · Documentation 9/20 Weakest · Reliability 4/20

Reliability
4/20
Security and permissions
7/20
Maintenance
9/20
Documentation
9/20
Setup experience
5/20
Why each score
Reliability 4/20
Only the README is supplied; there is no server manifest, implementation, MCP initialization or tool-list code, nor verifiable CI or tests. The README merely mentions MCP tool integration, so startup, handshake, tool consistency, and error handling cannot be verified. The score is conservatively limited by the static evidence.
Security and permissions 7/20
The README claims that file and shell operations are restricted to the workspace and that path traversal and dangerous commands are blocked. However, no implementation evidence covers the permission model, credential handling, network data flows, or confirmation for dangerous actions. Autonomous agents and external models/services may have side effects, so only partial credit is justified.
Maintenance 9/20
The repository is stated to be non-archived, uses the MIT license, and exposes GitHub issues, documentation, Discord, a changelog, and contribution guidance. These indicate support and maintenance channels, but there is no evidence of commit history, release cadence, dependency updates, a security-response process, or accountable ownership, so points are deducted.
Documentation 9/20
The README provides an installation command, Python quickstart examples, an OCR example, documentation/API/example links, and community support links. It does not document MCP server configuration, tool parameters and limits, authentication, cost, troubleshooting, or reviewable server behavior, preventing a near-full score.
Setup experience 5/20
A Python package installation command and ordinary agent examples are provided, but there is no MCP server manifest, launch command, transport information, mainstream-client configuration, or end-to-end connection procedure. MCP is only mentioned as an integration direction, so setup remains difficult to verify.

Static review · not runListed 2026-07-29

Read the FMRS scoring method →

Fit and risk

What it can accessWrites / deletes local filesRuns commands or codeUses the network

Best for

  • Developers building AI agents in Python
  • Autonomous-agent projects requiring workspace-scoped file and shell operations
  • Projects needing custom tools or OCR capabilities

Not for

  • Users seeking a ready-made MCP server tool catalog
  • Users requiring verified transport, authentication, or client configuration details
  • Projects that need unrestricted file or shell operations without an isolated environment

Required permissions

  • Autonomous-agent file and shell operations are restricted to the specified workspace
  • The README does not document additional MCP-server permissions, authentication, or external-service access requirements

Risks and side effects

  • Autonomous agents perform file and shell operations; although the README says path traversal and dangerous commands are blocked, the workspace scope should still be reviewed
  • A sandbox provider is described as a next step; the README mentions E2B but does not specify security settings
  • The supplied material does not establish that the repository itself exposes a directly connectable MCP server

Setup

Before you start

Runtime：Python

The README documents installing the Python package with uv pip install upsonic or pip install upsonic. It does not provide installation or client configuration steps for an MCP server.

Check that it works

After `uv pip install upsonic`, run `from upsonic import Agent, Task` and call agent.print_do(task); a normal returned result confirms the install.

Troubleshooting

  1. Confirm that the `upsonic` Python package is installed
  2. Check that the configured workspace path is correct and accessible
  3. Consult the Upsonic documentation, quickstart, and examples
  4. For MCP integration, verify the specific MCP tool and its configuration source

Things to try

Once connected, you can ask your AI assistant things like:

  • Create an Upsonic AutonomousAgent to analyze server logs and detect anomaly patterns
  • Write a Upsonic traditional Agent example using a @tool-decorated sum_tool function
  • Extract text from invoice.pdf using EasyOCREngine

Use cases

Building autonomous AI agents
Building traditional agent systems
Adding custom tools
Using MCP tools to connect agents to external data sources and services
OCR and document processing

Overview

Upsonic is a Python framework for building autonomous AI agents and traditional agent systems. Its README mentions integrating MCP tools, but provides no manifest, tool definitions, transport details, or client configuration showing that this repository is an MCP server.

Similar servers

Context7 80 · B

Upstash's official server providing up-to-date third-party library docs for AI coding assistants

★ 63.3k · Tools 2 Compare with this →
Compare with other MCP servers

Source revision 101f0313b0dd Data synced 2026-10-19 Read the FMRS scoring method