Open-source Computer Use Agent that works for Any App (GUI or CLI) by Any Model (Vision or Text).

Enikk is an AI agent that watches your screen, understands what it sees, and operates any desktop application — autonomously. It learns from experience, extracts reusable skills, and gets smarter every time you use it.

After each task, Enikk automatically reviews what happened and extracts reusable skills into persistent memory. The agent gets smarter the more you use it.

Enikk uses a multi-layer perception system (YOLO + OCR + VLM) that minimizes AI costs. Most operations never need expensive vision models — saving you money.

Configure everything through the Web Dashboard. No programming, no scripts, no YAML editing — just point-and-click setup.

Control multiple applications simultaneously. Enikk auto-discovers windows and can switch between apps during a task.

Control Enikk from your phone through chat. Send a task via QQ, and Enikk executes it on your PC while you're away.

Watch every step in the Web Dashboard: screenshots, detected elements, reasoning, and actions. No black boxes — complete transparency.

Runs entirely on your machine. Your data, screenshots, and credentials never leave your computer (except for AI API calls).

Model Configuration - Point-and-click setup

IM Configuration - Connect QQ Bot for remote control

Real-world Demo - Enikk controlling Netease Cloud Music
▶ Watch demo on YouTube

Join us for questions, feedback, and updates:

Discord                               QQ Group (627417852)

• Issues: Report bugs & request features

Data flow:

1. Capture — Take a screenshot of the target application window
2. Parse — YOLO detects UI elements, OCR reads text, results normalized to [0,1000] coordinates
3. Decide — LLM analyzes structured data and decides what to do
4. Act — Click, type, or wait based on the decision

1. Download the latest release from GitHub Releases
2. Unblock — Right-click the zip → Properties → Check "Unblock" → OK
3. Extract the zip file (see INSTALL.txt in the extracted folder for details)
4. Run enikk.exe
5. Open — the dashboard opens automatically in a native window

That's it. No installation, no Python, no dependencies.

Enikk can control any Windows desktop application:

# Install Git LFS first (for model weights)
git lfs install
git clone https://github.com/gtt116/enikk.git
cd enikk
uv venv --seed
uv pip install -e .
enikk

# Debug build (with console)
.\build.bat

# Release build (without console)
.\build.bat --release

Generated files:

• Debug mode: dist/enikk-debug/enikk-debug.exe
• Release mode: dist/enikk/enikk.exe and dist/enikk.zip

Enikk stands on the shoulders of giants:

Enikk's own code is MIT License — see LICENSE.

This project is intended for learning and research purposes only. Do NOT use this project in violation of any laws, game terms of service, or for any illegal purposes. Users bear full responsibility for any consequences arising from the use of this project.

MIT License — see LICENSE for details.