explaingit

ferroxlabs/wayland

Analysis updated 2026-05-18

137TypeScriptAudience · developerComplexity · 3/5LicenseSetup · moderate

TLDR

A local desktop hub that coordinates multiple AI coding assistants like Claude Code and Gemini CLI, with persistent memory and multi-agent teamwork.

Mindmap

mindmap
  root((Wayland))
    What it does
      AI assistant hub
      Cowork long running goals
      Persistent memory
      Multi agent teams
    Tech stack
      Rust engine
      Local databases
      Ollama support
    Use cases
      Coordinate coding tools
      Automate recurring tasks
      Run agents offline
    Audience
      Developers
      AI power users

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Coordinate multiple AI coding assistants from a single local interface instead of switching tools.

USE CASE 2

Assign a long-running goal that Wayland keeps working on across multiple sessions.

USE CASE 3

Assemble a team of specialized AI agents managed by one coordinator for larger jobs.

What is it built with?

RustTypeScriptOllama

How does it compare?

ferroxlabs/waylandmodelstudioai/cliamet2901/asterdex-trading-bot
Stars137137136
LanguageTypeScriptTypeScriptTypeScript
Setup difficultymoderateeasymoderate
Complexity3/52/54/5
Audiencedeveloperdevelopergeneral

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · moderate Time to first run · 30min

Windows and macOS builds are not code signed, so users must click through a security warning on first launch.

AGPL-3.0, a copyleft license that requires sharing source code, including for network-hosted modifications.

In plain English

Wayland is a desktop application that acts as a central hub for running and coordinating multiple AI coding assistants on your own computer. Instead of switching between separate AI tools like Claude Code, Gemini CLI, or Codex, you interact with all of them through a single interface. Your API keys, files, and shell access stay on your machine and never pass through an external server. The core feature is called Cowork: you describe a goal in plain language and Wayland keeps working toward it across multiple sessions. It maintains a persistent memory stored in five local databases so it remembers where the project stands even after you close the app. For larger jobs, it can assemble a group of specialized AI agents that share a common workspace, with one acting as a manager coordinating the others until the task is complete. Beyond conversation, Wayland can read and write actual files on your disk, run shell commands inside a security sandbox appropriate for your operating system, and execute recurring tasks on a schedule. The 47 MB engine is built in Rust and includes over 2,000 built-in capabilities across 177 workflows. It can route tasks to whichever AI model fits the work best, or run entirely offline using a local model provider like Ollama. The software is free and open source under the AGPL-3.0 license. Prebuilt installers are available for macOS (both Apple Silicon and Intel), Windows, and Linux. The Windows and macOS builds currently lack code signing, so first-time users will see a security warning they need to click through manually. A self-hosted cloud mode that lets you reach the agent from a phone is listed as shipping soon but is not yet available. The engine can also rewrite and re-score its own internal skill prompts against an evaluation process, which the README describes as making the agent sharper the more you use it. The project targets developers who want one consistent place to coordinate multiple AI tools without giving up local file access or paying for an additional subscription.

Copy-paste prompts

Prompt 1
Explain how Wayland's Cowork feature keeps working on a goal across multiple sessions.
Prompt 2
How do I install Wayland on macOS and get past the unsigned-app security warning?
Prompt 3
What does it mean that Wayland can run entirely offline with a local model like Ollama?
Prompt 4
How does Wayland assemble a team of AI agents to complete a larger task?

Frequently asked questions

What is wayland?

A local desktop hub that coordinates multiple AI coding assistants like Claude Code and Gemini CLI, with persistent memory and multi-agent teamwork.

What language is wayland written in?

Mainly TypeScript. The stack also includes Rust, TypeScript, Ollama.

What license does wayland use?

AGPL-3.0, a copyleft license that requires sharing source code, including for network-hosted modifications.

How hard is wayland to set up?

Setup difficulty is rated moderate, with roughly 30min to a first successful run.

Who is wayland for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.