Ywza offers model-driven PC control with paired remote access
state: seedheat: mediumuncertainty: mediumrelevance: mediumcomputer-use-agent remote-controlalielmorsy
What is this?
Ywza is a GitHub project by alielmorsy, introduced in a Show HN post dated Nov. 8, 2026. Its project page describes a background agent that lets a compatible AI model inspect windows, click and type in apps, and use open terminals on a PC. It supports same-machine use and pairing with another device via a one-time code; it says paired devices can connect through public relays without exposing the PC directly to the internet. Users supply a model compatible with the OpenAI chat API, including local models. The project says Windows works and Linux support is limited to X11. The source documents build and setup instructions, but provides no independent evaluation, adoption data, or security audit; its “open-source” status is not verifiable from the supplied text alone.
Why it matters
Ywza brings model-driven computer use to a user-managed PC and adds paired remote access, potentially letting people combine their preferred model—including a local one—with desktop workflows. That is a practical extension of agent capability, but the available evidence is a project description and demos, not proof of reliability, security, or broad use. Public relays are part of the described remote connection path, so “without exposing the PC directly” should not be read as “no network or third party involved.”
Current assessment
New: alielmorsy presented Ywza on Show HN on Nov. 8, 2026; the linked GitHub page provides build and quick-start instructions. According to the project, the agent exposes window elements to the model as numbered lists, can request screenshots, and can control apps and open terminals. Users configure a compatible model; local options such as Ollama and LM Studio are named. Remote pairing uses a single-use six-digit code that expires after 10 minutes. The project says paired devices may connect across networks through public relays, while own-relay support is only planned. It reports Windows support and Linux support for X11 only. These are developer claims; supplied evidence does not establish security properties beyond the stated connection design, nor provide independent testing, adoption, or a confirmed license. Research follow-up: inspect the repository license, relay and pairing implementation, and how model/API traffic and credentials are handled before making stronger privacy or security claims.
Measured heat
now 0 pts/hpeak 2 pts/hcomments 0/hpeers p21momentum: steady2 platformsage 120h
points/hour across evidence · reading as of 2026-11-13 09:25:35.504452+11:00 · deterministic, not a model opinion
How the heat travelled
pace: p10 vs 1908 stories at the 96h mark (now 120h old) — behind 0sql-semantics-to-sql-engine (0.5x)
Evidence (2) — ⭐ canonical anchor
| source | object | author | score | comments |
| 🟧 hn | Show HN: Ywza, a computer-use agent for your own PC, drivable from anywhereRetrieved article excerptOpen article · Retrieved 2026-11-07T23:22:43.732955+00:00 # ywza
**Let an AI model use your PC, from anywhere.**
[Demo](https://github.com/alielmorsy/ywza#demo) •
[Build](https://github.com/alielmorsy/ywza#build) •
[Quick start](https://github.com/alielmorsy/ywza#quick-start) •
[Pairing](https://github.com/alielmorsy/ywza#control-it-from-another-device) •
[MCP](https://github.com/alielmorsy/ywza#use-it-from-an-mcp-client) •
[Linux](https://github.com/alielmorsy/ywza#linux)
---
ywza lets an AI model use your computer the way you do. It can look at your windows, click and type in any app, and work
in the terminals you already have open. You can talk to it from the same machine or from another one, and nothing has to
be exposed to the internet to make that work.
It runs as a small background agent on the PC you want to control. Other devices connect to it directly, peer to peer,
after you pair them once with a 6-digit code. There is no account and no server to host. You bring your own model:
anything that speaks the OpenAI chat API works, including local ones like Ollama and LM Studio.
Instead of sending the model a screenshot at every step, ywza describes each window as a numbered list of elements. That
keeps the token count low and the model's reasoning focused. It can still ask for a screenshot when it needs to see
something.
ywza works perfectly on Windows. Linux works on X11.
## Demo
| | |
| --- | --- |
| hello\_world.mp4 Opening Notepad, typing Hello World and moving the window to monitor 0 | open\_reddit.mp4 Opening Reddit in Chrome |
| terminal\_access.mp4 Typing Hello World into an empty terminal, then starting Claude in it | |
## Build
You need a recent [Rust toolchain](https://rustup.rs). On Windows, the MSVC build tools that rustup asks for are enough.
```
git clone https://github.com/alielmorsy/ywza
cd ywza
cargo build --release
```
This produces two programs in `target/release`:
| Program | What it is |
| --- | --- |
| `ywza-agent` | The background service. Runs on the machine you want to control. |
| `ywza` | The command line client. Runs wherever you give the instructions. |
Put both on your PATH, or install them with cargo:
```
cargo install --path crates/ywza-agent
cargo install --path crates/ywza-cli
```
### OCR
Some apps draw their own text and expose nothing to Windows accessibility. For those, build the agent with an OCR
engine. You can compile several in and pick one in the config.
| Feature | Engine | Platform |
| --- | --- | --- |
| `ocr-windows` | Built-in Windows OCR | Windows |
| `ocr-paddle` | PaddleOCR through a bundled ONNX Runtime | Windows, Linux |
| `ocr-paddle-dynamic` | PaddleOCR, ONNX Runtime loaded at startup | Linux, older distros |
```
cargo build --release -p ywza-agent --features ocr-windows
```
PaddleOCR models download on first use. `ocr-paddle-dynamic` replaces `ocr-paddle`, so don't enable both. It needs
`libonnxruntime.so` next to the agent binary, or at the path in `ORT_DYLIB_PATH`.
## Quick start
Start the agent:
```
ywza-agent run
```
The first launch creates your config folder and the agent's identity. Point it at a model:
```
ywza config set-llm --base-url https://api.openai.com/v1 --model <model> --api-key <key>
```
Then give it something to do:
```
ywza prompt "Open Notepad and write hello"
```
## Control it from another device
On the machine you want to control, ask the running agent for a pairing code:
```
ywza-agent pair
```
On the device you want to control it from:
```
ywza pair 123456 --name desktop
ywza -d desktop prompt "Take a screenshot and tell me what's open"
```
The code works once and expires after 10 minutes. On Windows, if the connection times out, run
`dist/windows-firewall.bat` as administrator with the path to `ywza-agent.exe`.
## Use it from an MCP client
The MCP server gives any MCP client control of this PC and of every device you have paired, through one server. The model
picks the device per call, so a single chat can work on your desktop and a remote machine side by side.
Paired devices don't need to be on the same network. They connect through public relays, so a laptop on a coffee shop
wifi can drive a PC at home. Own relay support is planned, so you will be able to run your own instead of the public
ones.
Register it straight from the repo, no install step. Replace `<repo>` with the folder you cloned ywza into:
```
claude mcp add ywza -- uv run --quiet --project <repo>/python ywza-mcp --cli <repo>/target/release/ywza.exe
```
On Linux the last path is `<repo>/target/release/ywza`. `--cli` is how the server reaches your paired devices, so
it can be left out if `ywza` is already on your PATH or you only control this PC. Any other MCP client can run the same
command.
## Linux
Only X11 sessions are supported. `dist/ywza-agent.service` runs the agent on the login screen and in every user
session. Edit the `ExecStart` path, then follow the comments at the top of the file. | alielmorsy19 | 1 | 0 |
| 🟧 echo.github ⭐ | Ywza describes an agent that lets an AI model inspect windows, click and type, and use open terminals on a PC; paired devices connect peer t | alielmorsy | — | — |
Interpretation history
2026-11-07T23:23:07Z
audience grounding: medium — Ywza brings model-driven computer use to a user-managed PC and adds paired remote access, potentially letting people combine their preferred model—including a local one—with desktop workflows. That is a practical extension of agent capability, but the available evidence is a project description and demos, not proof of reliability, security, or broad use. Public relays are part of the described remote connection path, so “without exposing the PC directly” should not be read as “no network or third party involved.”
Decision trace
- 11-08 10:24attention_candidatecreate
- 11-08 10:23groundYwza brings model-driven computer use to a user-managed PC and adds paired remote access, potentially letting people combine their preferred model—including a local one—with desktop workflows. That is
- 11-08 10:22createThe repository documents a usable remote computer-control tool with a distinct pairing and connectivity design, though the supplied evidence is a single low-engagement announcement.