explaingit

wavetermdev/waveterm

📈 Trending★ 20,511GoAudience · developerComplexity · 4/5ActiveLicenseSetup · moderate

TLDR

A modern terminal app that puts an AI assistant, file editor, and web browser side-by-side with your command line, with support for remote SSH sessions and local AI models.

Mindmap

mindmap
  root((repo))
    What it does
      AI-powered terminal
      Multi-pane layout
      File editor built-in
      Web browser pane
    Key features
      Context-aware AI chat
      Durable SSH sessions
      Multiple AI providers
      Local model support
    Use cases
      Debug code faster
      Manage remote servers
      Automate workflows
      Learn from errors
    Tech stack
      Go backend
      Native desktop app
      OpenAI Claude Gemini
      Ollama LM Studio
    Audience
      Developers
      System admins
      DevOps engineers

Things people build with this

USE CASE 1

Debug errors in your code by asking the AI assistant to read your terminal output and suggest fixes.

USE CASE 2

Connect to remote servers over SSH and keep sessions alive even if your network drops or you restart the app.

USE CASE 3

Write and edit files in the built-in editor while running commands in the terminal pane next to it.

USE CASE 4

Pipe command output directly into the AI chat to get instant explanations or suggestions without leaving the terminal.

Tech stack

GoOpenAIClaudeGeminiOllamaLM Studio

Getting it running

Difficulty · moderate Time to first run · 30min

Requires Go installation and either an API key (OpenAI/Claude/Gemini) or local LLM setup (Ollama/LM Studio) to function.

Use freely for any purpose including commercial. Keep the notice and disclose changes to the patent grant.

In plain English

Wave Terminal is an open-source terminal application for macOS, Linux, and Windows that integrates an AI assistant directly into your command-line workspace. A terminal is the text-based interface developers use to run programs, manage files, and connect to remote servers. Wave takes that familiar environment and adds a flexible, panel-based layout where terminal windows, a built-in file editor, a file previewer, a web browser pane, and an AI chat widget can all sit side by side and share context with each other. The AI assistant feature, called Wave AI, is context-aware: it can read what is currently on your terminal screen, help you debug errors, read and write files, and receive piped command output directly from the command line. It supports major AI providers (OpenAI, Claude, Gemini, Azure, Perplexity) using your own API keys, and also works with locally-run models through Ollama or LM Studio, so no account with a cloud provider is strictly required. One notable feature is durable SSH sessions: when you connect to a remote server over SSH, Wave keeps that session alive through network drops and application restarts, automatically reconnecting. There is also a companion command-line tool called wsh that lets you manage your Wave workspace, copy files between local and remote machines, and pipe output into the AI from any terminal. Wave is built in Go and is aimed at developers and system administrators who want a more modern, integrated terminal experience without switching entirely to a GUI tool. It installs as a native desktop application and requires macOS 11+, Windows 10+, or a reasonably modern Linux distribution.

Copy-paste prompts

Prompt 1
I'm getting a permission denied error when I try to SSH into my server. Can you help me debug this using Wave Terminal's AI assistant?
Prompt 2
Show me how to set up Wave Terminal with my local Ollama model so I can use AI without paying for cloud API keys.
Prompt 3
How do I use the wsh command-line tool to copy files between my local machine and a remote server through Wave Terminal?
Prompt 4
I want to pipe the output of a failing build command into Wave Terminal's AI chat to get debugging suggestions. How do I do that?
Prompt 5
Set up Wave Terminal with multiple AI providers (OpenAI and Claude) so I can switch between them in the same workspace.
Open on GitHub → Explain another repo

Generated 2026-05-18 · Model: sonnet-4-6 · Verify against the repo before relying on details.