Back to Registry
Environment Variables (backend
View Author Profile
Official Verified
Openyak Desktop Agent
Skill by adisinghstudent
skill-install — Terminal
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/adisinghstudent/openyak-desktop-agentOr
---
name: openyak-desktop-agent
description: OpenYak local-first AI agent desktop app with 100+ models, 16+ built-in tools, and MCP support
triggers:
- set up OpenYak desktop agent
- configure OpenYak with my API key
- use OpenYak for file automation
- build a skill or tool for OpenYak
- connect OpenYak to local models
- OpenYak agent modes and tools
- integrate MCP server with OpenYak
- OpenYak backend API development
---
# OpenYak Desktop Agent
> Skill by [ara.so](https://ara.so) — Daily 2026 Skills collection.
OpenYak is a 100% local-first, open-source desktop AI assistant that runs entirely on your machine. It supports 100+ AI models via OpenRouter, 16+ built-in tools (file I/O, bash, web fetch, glob/grep), 7 specialized agent modes, MCP server integration, and a secure remote tunnel for mobile access — all without uploading data to the cloud.
---
## Installation
### End-User (Desktop App)
1. Download the installer from [https://open-yak.com/download/](https://open-yak.com/download/) (Windows or macOS).
2. Launch the app and connect a model (free tier: 1M tokens/week, or bring your OpenRouter API key).
### Developer Setup
The project has two parts: `frontend/` (Electron/UI) and `backend/` (Python).
#### Backend
```bash
cd backend
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env # then edit .env
python main.py
Frontend
cd frontend
npm install
npm run dev # development mode with hot reload
npm run build # production build
npm run package # create distributable installer
See frontend/README.md and backend/README.md for full setup details.
Configuration
Environment Variables (backend .env)
# Model provider — use OpenRouter or your own key
OPENROUTER_API_KEY=$OPENROUTER_API_KEY
# Optional: direct provider keys
OPENAI_API_KEY=$OPENAI_API_KEY
ANTHROPIC_API_KEY=$ANTHROPIC_API_KEY
# Local model endpoint (e.g. Ollama)
LOCAL_MODEL_BASE_URL=http://localhost:11434/v1
# App settings
DATA_DIR=~/.openyak/data
LOG_LEVEL=INFO
Connecting a Model in the UI
- Open Settings → Models.
- Choose a provider: OpenRouter, OpenAI, Anthropic, or Local (Ollama/LM Studio).
- Paste your API key or leave blank for free-tier OpenRouter models.
- Select a default model (e.g.
anthropic/claude-opus-4.6ordeepseek/deepseek-v3.2).
Agent Modes
OpenYak ships 7 specialized agents:
| Mode | Purpose |
|---|---|
| Build | Code generation, project scaffolding |
| Plan | Multi-step task decomposition |
| Explore | File system and data investigation |
| Write | Document drafting and editing |
| Analyze | Spreadsheet / CSV data analysis |
| Automate | Office workflow automation |
| Chat | General conversation |
Metadata
AI Skill Finder
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skill Add to Configuration
Paste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-adisinghstudent-openyak-desktop-agent": {
"enabled": true,
"auto_update": true
}
}
}Safety NoteClawKit audits metadata but not runtime behavior. Use with caution.
Related Skills
Planning With Files Manus Workflow
Skill by adisinghstudent
adisinghstudent 3809
claude-better-cli
Compatibility-first Claude CLI reimplementation with faster startup, lower memory, and drop-in command compatibility
adisinghstudent 3809
mirofish-offline-simulation
Fully local multi-agent swarm intelligence simulation engine using Neo4j + Ollama for public opinion, market sentiment, and social dynamics prediction.
adisinghstudent 3809
ghostling-libghostty-terminal
Build minimal terminal emulators using the libghostty-vt C API with Raylib for windowing and rendering
adisinghstudent 3809
Obra Superpowers Agentic Workflow
Skill by adisinghstudent
adisinghstudent 3809