UFO vs wshobson/agents
Which should you use? A side-by-side look at what each one does, who it suits, what it costs, how hard it is to set up, and where each falls short.
Open source · MIT
Open-source Windows agent that uses an LLM to operate desktop apps, and a multi-device orchestration layer cal
Open source · MIT
A collection of plugins, agents, skills, and commands that run across several AI coding tools.
The short answer
Choose UFO if you need…
- Developers experimenting with LLM-driven Windows desktop automation
- Researchers studying multi-agent task orchestration
- People who want to automate repetitive tasks in Windows apps and are comfortable with setup
- Teams exploring cross-device workflows, using the Galaxy framework
Choose wshobson/agents if you need…
- Developers who already use one of the listed AI coding tools and want ready-made specialist agents
- Python or JavaScript/TypeScript developers looking for testing, packaging, or project setup guidance
- Teams wanting a shared set of review and security-scanning workflows across different tools
- People who want to install only individual skills without the full plugin set
Side-by-side
| UFO | wshobson/agents | |
|---|---|---|
| Type | AI model / agent | Dev tool |
| Price | Free and open source | Free and open source |
| Licence | MIT | MIT |
| GitHub stars | 10.2k | 42.1k |
| Setup difficulty | Advanced | Beginner |
| How you get it | Clone and build from source | Install with a package manager |
| Main language | Python | Python |
| Last code update | 7 days ago | 4 days ago |
What each one is
UFO
UFO³ is a Microsoft research project for building AI agents that control Windows computers and coordinate work across several devices. Its older part, UFO², runs on a single Windows machine and uses a language model to read the screen and act on apps through clicks and API calls. The newer Galaxy part splits a request into a graph of tasks and assigns them to different devices, such as Windows, Linux, or Android, as described in the README.
wshobson/agents
This project is a shared library of prompts-as-plugins for AI coding assistants. It contains 94 plugins with specialist agents, skills (reference guidance an agent loads when relevant), and commands (named workflows) covering areas like Python, JavaScript, code review, testing, infrastructure, and security. One Markdown source is adapted to Claude Code, OpenAI Codex CLI, Cursor, OpenCode, Antigravity CLI, GitHub Copilot, and Pi.
Best for
UFO
- Developers experimenting with LLM-driven Windows desktop automation
- Researchers studying multi-agent task orchestration
- People who want to automate repetitive tasks in Windows apps and are comfortable with setup
- Teams exploring cross-device workflows, using the Galaxy framework
wshobson/agents
- Developers who already use one of the listed AI coding tools and want ready-made specialist agents
- Python or JavaScript/TypeScript developers looking for testing, packaging, or project setup guidance
- Teams wanting a shared set of review and security-scanning workflows across different tools
- People who want to install only individual skills without the full plugin set
Strengths
UFO
- Open source under the MIT license with a public paper and documentation site
- Two tiers: UFO² is described as simpler and stable, while Galaxy adds multi-device orchestration
- Combines GUI actions with API calls, which the README says can reduce LLM calls
wshobson/agents
- Covers seven coding tools from one shared source of components
- Large catalog: 94 plugins, 202 agents, 185 skills, and 105 commands according to the README
- Offers a skills-only install route that does not require the full plugin set
Not ideal for
UFO
- Non-technical users wanting a ready-made consumer app, since the README describes setup and configuration steps
- Anyone needing a stable, fully tested production system, because the README labels Galaxy as in active development
- Users who need macOS-only automation, since the README lists Windows, Linux, and Android as supported device types
wshobson/agents
- People who do not use any supported AI coding tool, since the components only run inside those harnesses
- Anyone expecting the scores from the plugin-eval tool to prove a task will succeed, because the README says the LLM judge and Monte Carlo layers are experimental and not validated against human labels
- Users wanting a fully stable, versioned release, since the README lists no latest release
Trade-offs
UFO
- Depends on a language model and its API access, which the README discusses but this summary does not detail
- Galaxy setup is described as moderate in difficulty, and it is still in active development
- Primary focus is Windows, so cross-platform reach depends on the Galaxy device options
wshobson/agents
- The README notes that available components depend on the installation route and each harness's limits, so features differ by tool
- Some features (LLM judge, Monte Carlo evaluation) are experimental and may bill your Anthropic API account if ANTHROPIC_API_KEY is set
- Pi agents need a separate subagent extension, and the Codex native and generated routes behave differently
Getting started
UFO
Clone the repository
git clone https://github.com/microsoft/UFO.gitwshobson/agents
Claude Code: register marketplace (run inside Claude Code)
/plugin marketplace add wshobson/agentsClaude Code: install Python plugin (run inside Claude Code)
/plugin install python-development@claude-code-workflowsFrequently asked questions
What is the difference between UFO and wshobson/agents?
UFO: Open-source Windows agent that uses an LLM to operate desktop apps, and a multi-device orchestration layer cal. wshobson/agents: A collection of plugins, agents, skills, and commands that run across several AI coding tools. UFO suits developers experimenting with llm-driven windows desktop automation, while wshobson/agents suits developers who already use one of the listed ai coding tools and want ready-made specialist agents.
Are UFO and wshobson/agents free?
UFO: Free and open source. wshobson/agents: Free and open source.
Which is easier to set up, UFO or wshobson/agents?
wshobson/agents is easier to get started with (rated beginner); UFO is rated advanced and usually needs more setup.
Is UFO or wshobson/agents more popular?
wshobson/agents has more GitHub stars (42.1k) than UFO (10.2k). Popularity isn't the same as fit — check which one matches your use case above.
What are other alternatives to UFO and wshobson/agents?
Other options people compare them with include Self-Operating Computer, RD-Agent, DeerFlow, SWE-agent, Open SWE.
Keep comparing
Built from our catalogue entries for both products: open-source details come from each project's README (AI-assisted, refreshed weekly) and GitHub data; AI-tool details are editorial. Spotted something wrong? Let us know from either product's page.