> ## Documentation Index
> Fetch the complete documentation index at: https://docs.yambr.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Introduction

> What Open Computer Use gives you, what makes it different, and why you'd pick it over Claude.ai, OpenAI Operator, or open-terminal.

## In one sentence

Open Computer Use is a **self-hosted MCP server** that gives any LLM a full Ubuntu sandbox — one isolated Docker container per chat session — with a live browser, an interactive terminal, code execution in bash/Python/Node/Java, and 13+ document skills. Yambr runs a managed instance you reach as an MCP tool server at `https://api.yambr.com/mcp/computer_use`; artifacts are served from `cu.yambr.com`. You bring your own model provider — Yambr doesn't resell inference.

## What you can do with it

| Capability                 | Example prompt                                                                                                          |
| -------------------------- | ----------------------------------------------------------------------------------------------------------------------- |
| Create a polished document | *"Write me a 5-page technical RFC in .docx with headings, tables, and tracked changes."*                                |
| Analyse a spreadsheet      | *"Open this .xlsx, pivot by region, and chart monthly revenue."*                                                        |
| Build a landing page       | *"Make a one-page site for a coffee shop. Use Tailwind. Show me the live preview."*                                     |
| Drive a real browser       | *"Log in to our staging site, click through the onboarding flow, and screenshot each step."*                            |
| Delegate to Claude Code    | *"Clone this repo, run the tests, and open a PR with the fix."* — the sub-agent takes over with an interactive terminal |

Every file the model creates gets a public URL that you can preview inline or download.

## Key differentiators

| Feature                | Open Computer Use                    | Claude.ai (Claude Code web)           | [open-terminal](https://github.com/open-webui/open-terminal) | OpenAI Operator  |
| ---------------------- | ------------------------------------ | ------------------------------------- | ------------------------------------------------------------ | ---------------- |
| Self-hosted            | Yes                                  | No                                    | Yes                                                          | No               |
| Any LLM                | Yes (OpenAI-compatible)              | Claude only                           | Any (via Open WebUI)                                         | GPT only         |
| Code execution         | Full Linux sandbox                   | Sandbox (Claude Code web)             | Sandbox / bare metal                                         | No               |
| Live browser           | CDP streaming (shared, interactive)  | Screenshot-based                      | No                                                           | Screenshot-based |
| Terminal + Claude Code | ttyd + tmux + Claude Code CLI        | Built-in (Claude Code web)            | PTY + WebSocket                                              | N/A              |
| Skills system          | 13 built-in (auto-injected) + custom | Built-in skills + custom instructions | Open WebUI native (text-only)                                | N/A              |
| Container isolation    | Docker (runc), per chat              | Docker (gVisor)                       | Shared container (OS-level users)                            | N/A              |

See the [full comparison](/reference/comparison) for caveats.

## What's inside the sandbox

| Category      | Tools                                                   |
| ------------- | ------------------------------------------------------- |
| **Languages** | Python 3.12, Node.js 22, Java 21, Bun                   |
| **Documents** | LibreOffice, Pandoc, python-docx, python-pptx, openpyxl |
| **PDF**       | pypdf, pdf-lib, reportlab, tabula-py, ghostscript       |
| **Images**    | Pillow, OpenCV, ImageMagick, sharp, librsvg             |
| **Web**       | Playwright (Chromium), Mermaid CLI                      |
| **AI**        | Claude Code CLI, Playwright MCP                         |
| **OCR**       | Tesseract (configurable languages)                      |
| **Media**     | FFmpeg                                                  |
| **Diagrams**  | Graphviz, Mermaid                                       |
| **Dev**       | TypeScript, tsx, git                                    |

<Note>
  Skills are **model-agnostic**. Write a skill once with Claude Code in the terminal and then use it with any model in the chat — the auto-injected system prompt exposes every enabled skill to every model.
</Note>

## Next

<CardGroup cols={2}>
  <Card title="Architecture" icon="diagram-project" href="/architecture">
    How a request flows from the chat through the MCP server into the sandbox.
  </Card>

  <Card title="Yambr Platform" icon="building" href="/platform/overview">
    The four services that make up Yambr and what each is for.
  </Card>
</CardGroup>