Wave Terminal: A Cross-Platform Workspace That Merges Terminal, Editor, and AI
Wave combines a terminal, editor, browser, and AI assistance in one cross-platform workspace.
At a glance
- What is it?
- Wave Terminal combines a terminal, editor, browser, and AI assistant into one desktop app for macOS, Linux, and Windows. It targets developers who want durable SSH sessions and context-aware AI help without leaving the command line.
- Who is it for?
- Adopt Wave Terminal if you manage remote servers over flaky connections and want an AI assistant that reads your terminal context, or if you prefer a BYOK model with no account lock-in. Avoid it if you need a minimal terminal or if your Linux environment predates glibc 2.28.
- Can I use it commercially?
- Yes. Apache-2.0 is a permissive licence: you can use, modify and sell software built on it, as long as you keep its copyright and licence notices.
- Is it still maintained?
- Yes. The repository last received commits 6 days ago.
- What is it written in?
- Mainly Go, according to GitHub's language statistics.
Answers come from the project's GitHub data, last synced on September 15, 2026, and from our analysis. They are not legal advice.
DEEP OPEN-SOURCE ANALYSIS
The Problem: Fragmented Remote Workflows
Developers who work on remote servers often juggle a terminal, a separate editor, a browser for documentation, and an AI chat window. Each tool has its own context, so debugging a failed deployment means copying output from one app to another. Wave Terminal attacks this by putting all four into a single window. The README describes it as an open-source, AI-integrated terminal for macOS, Linux, and Windows. The target user is someone who lives in a terminal but wants graphical editing and AI assistance without switching apps. It is not a replacement for a full IDE, but it aims to reduce context switching for remote operations.
How Wave Works: Blocks, wsh, and AI Context
Wave's core architecture is a flexible drag-and-drop interface that organizes terminal blocks, editors, web browsers, and AI assistants. Each block can be toggled to full-screen and back, which helps when you need to focus on one task. The `wsh` command system is the glue: it lets you manage your workspace from the CLI and share data between terminal sessions. For example, `wsh file` copies and syncs files between local and remote SSH hosts. The AI assistant, Wave AI, reads terminal output and scrollback, analyzes widgets, and performs file operations with automatic backups and user approval. That context-awareness is the key differentiator: the AI sees what you see, not just a prompt you typed.
Durable SSH and Remote File Editing
A standout feature is durable SSH sessions. The README states that remote terminal sessions survive connection interruptions, network changes, and Wave restarts with automatic reconnection. This is a practical benefit for anyone on unstable Wi-Fi or who regularly suspends a laptop. The built-in editor handles remote files with syntax highlighting, and the file preview system supports markdown, images, video, PDFs, CSVs, and directories. That means you can inspect a remote log file or image without downloading it. The one-click remote connections give full terminal and file system access, which is convenient but also a security consideration: anyone with access to your Wave workspace could reach those hosts.
Getting Started: Installation and Requirements
Installation is platform-specific, with instructions linked from the README to docs.waveterm.dev/gettingstarted. You can also download directly from www.waveterm.dev/download. Minimum requirements are reasonable: macOS 11 or later (arm64, x64), Windows 10 1809 or later (x64), and Linux based on glibc-2.28 or later, such as Debian 10, RHEL 8, or Ubuntu 20.04. The WSH helper has looser Linux kernel requirements (2.6.32 for x64, 3.1 for arm64), but the main app needs a modern glibc. If you are on an older enterprise Linux distribution, this could be a blocker. There are no accounts required, and you bring your own API keys for AI providers.
AI Integration: BYOK and Local Models
Wave AI supports OpenAI, Claude, Gemini, Azure, and Perplexity, plus local models via Ollama and LM Studio. The README emphasizes that it works with any AI model, as long as you bring your own API keys. There is a free beta with included AI credits, but that is temporary. The CLI integration via `wsh ai` lets you pipe output or attach files directly from the command line, which is a nice touch for scripting. However, command execution is listed as 'Coming Soon', so the AI can read and edit files but cannot run commands yet. That is a real limitation: the assistant is a helper, not an autonomous agent.
Limitations and Trade-offs
The most obvious limitation is that Wave is a heavy desktop app, not a lightweight terminal emulator. If you prefer minimal tools like `alacritty` or `kitty`, Wave's browser and editor blocks may feel like bloat. The AI features depend on external API keys, which means you are either paying for usage or relying on local models that may be slower. The README does not mention any offline mode for the AI, so a network drop could disable that feature. Also, the file operations performed by Wave AI require user approval, which is good for safety but adds friction. Finally, the project is actively maintained, but the roadmap is not detailed in the README, so you should check the ROADMAP.md file for upcoming features.
Alternatives and Comparisons
The closest alternative in approach is a traditional terminal multiplexer like `tmux` combined with a separate editor like VS Code Remote SSH. That setup gives you durable sessions and remote editing, but it lacks the integrated AI and file previews. Another alternative is a full IDE like JetBrains or VS Code with built-in terminal and AI plugins. Those tools have richer editing and debugging, but they are heavier and less terminal-centric. Wave's difference is that it starts from the terminal and adds graphical layers, rather than starting from an IDE and adding a terminal. For AI, you could use a standalone tool like `shell-gpt` that pipes output to a model, but that does not have Wave's context awareness of widgets or file operations.
Maintenance and License
Wave Terminal is licensed under Apache-2.0, which is permissive for commercial use, but you should review the ACKNOWLEDGEMENTS.md file for dependencies. The project is not archived, and the last push was April 2026 with version v0.14.5, indicating active development. The README mentions a roadmap that is continuously updated, so maintenance appears ongoing. However, the AI features are in beta, and the free credits will not last forever. Upgrade cost is low in terms of license fees, but there is a learning curve for the wsh command system and the block-based interface. You should budget time to learn the keyboard shortcuts and workflow, especially if you are coming from a plain terminal.
Editorial conclusion
Adopt Wave Terminal if you manage remote servers over flaky connections and want an AI assistant that reads your terminal context, or if you prefer a BYOK model with no account lock-in. Avoid it if you need a minimal terminal or if your Linux environment predates glibc 2.28. Before adopting, verify that your AI provider is supported (OpenAI, Claude, Gemini, Azure, Perplexity, or any OpenAI-compatible local model) and test the wsh command system for your workflow. The project is actively maintained with recent releases, but check the roadmap for command execution, which is still listed as coming soon.
Community notes