Openyak
Open-source local-first AI agent for desktop work. No account, no telemetry: use local models with Ollama/Rapid-MLX or bring your own provider key.
What is Openyak?
Openyak is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to open-source local-first ai agent for desktop work. no account, no telemetry: use local models with ollama/rapid-mlx or bring your own provider key.
Open-source local-first AI agent for desktop work. No account, no telemetry: use local models with Ollama/Rapid-MLX or bring your own provider key.
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- Open-source local-first AI agent for desktop work. No accoun
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx openyakConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Openyak
OpenYak is a local-first, open-source AI agent for desktop work that runs on macOS, Windows, and Linux without requiring an OpenYak account, telemetry, or a hosted backend. It lets users attach real office files (DOCX, XLSX, PPTX, PDF, CSV) and work with local models through Ollama or Rapid-MLX (Apple Silicon), while optionally using BYOK cloud providers including Anthropic, OpenAI, Google, OpenRouter, and others. OpenYak functions as a full desktop agent with file read/write tool execution, long-context threading, a reusable artifact workspace, multi-agent task batches, and remote access via QR code and Cloudflare Tunnel. Developers and knowledge workers who want an AI agent that lives on their machine rather than behind a hosted workspace account use OpenYak as an alternative to cloud-only tools.
Prerequisites
- macOS, Windows, or Linux operating system (x64 or Apple Silicon)
- For local model inference: Ollama installed (https://ollama.com) or Rapid-MLX for Apple Silicon
- For cloud models: your own API key from a provider such as Anthropic, OpenAI, Google, or OpenRouter
- No OpenYak account required — the app is self-contained
- Optional: Node.js if building from source (monorepo uses Tauri v2, Rust, Next.js 15, FastAPI)
Download and install the OpenYak desktop application
Download the installer for your platform from the latest GitHub release or from open-yak.com/download. On macOS, open the .dmg and drag to Applications. On Windows, run the .exe installer. On Linux, install the .deb or .rpm package.
# Visit: https://github.com/openyak/openyak/releases/latest
# Or: https://open-yak.com/downloadConfigure your inference provider
Open OpenYak Settings and choose where inference runs. For offline local inference, select Ollama (auto-detected if running) or Rapid-MLX on Apple Silicon. For cloud models, select a provider and enter your BYOK API key.
Start a new conversation and attach a file
Click 'New Conversation', attach a real office file (DOCX, XLSX, PPTX, PDF, or CSV), and ask for a specific deliverable rather than a summary — for example, a brief, action plan, RACI, or follow-up email.
Use the artifact panel for reusable outputs
When OpenYak generates a structured output (Markdown brief, table, diagram, checklist), it opens in the right-side artifact panel so you can reference, export, or continue building on it in subsequent turns of the same thread.
Enable remote access for mobile use
In Settings, enable Remote Access. OpenYak creates a Cloudflare Tunnel and displays a QR code you can scan from a mobile device to send tasks to your desktop agent securely.
Openyak Examples
Client configuration
OpenYak is a standalone desktop app, not a traditional MCP server with a JSON config entry. If you want to integrate OpenYak's agent engine as an MCP server in another client, refer to the developer instructions for running the FastAPI backend on port 8000:
{
"mcpServers": {
"openyak": {
"command": "npx",
"args": ["openyak"]
}
}
}Prompts to try
Use OpenYak to turn real files into structured deliverables:
- "Read the files I uploaded and turn them into a concise team brief with three key takeaways, risks, owners, and next actions"
- "Analyze this spreadsheet for budget vs. actual variance and highlight the top three anomalies"
- "Synthesize these three documents into a board brief that reconciles all sources"
- "Create a RACI matrix from this project memo and save it as an artifact"
- "Continue from yesterday's thread and draft a follow-up email based on the action items we identified"Troubleshooting Openyak
Local models not detected or Ollama connection fails
Ensure Ollama is running ('ollama serve') before launching OpenYak. On Apple Silicon, verify Rapid-MLX is installed and a model is downloaded. OpenYak auto-detects local endpoints but requires the model server to be active at startup.
File attachment fails or office files produce empty results
OpenYak supports DOCX, XLSX, PPTX, PDF, and CSV. Ensure the file is not password-protected. Large files may take longer to parse — check the Linux-specific requirements in LINUX.md if running on Linux, as some system libraries may be needed for document parsing.
Frequently Asked Questions about Openyak
What is Openyak?
Openyak is a Model Context Protocol (MCP) server that open-source local-first ai agent for desktop work. no account, no telemetry: use local models with ollama/rapid-mlx or bring your own provider key. It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Openyak?
Follow the installation instructions on the Openyak GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Openyak?
Openyak works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Openyak free to use?
Yes, Openyak is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Openyak Alternatives — Similar Coding Agents Servers
Looking for alternatives to Openyak? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Openyak in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Openyak?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.