During his sabbatical, Will McGugan, maker of Rich and Textual( frameworks for making Textual User Interfaces (TUI)), put his UI skills to work to build Toad. Thenewly publicly released tool aims to provide a unified, "beautiful" GUI for multiple coding agents in your terminal, accessible via the same tool via the Agent Communication Protocol (ACP).
McGugan, who dedicated his last couple of years to building frameworks that make terminal applications more usable and appealing to users, believes that applications from companies in the AI space are less appealing because t…
During his sabbatical, Will McGugan, maker of Rich and Textual( frameworks for making Textual User Interfaces (TUI)), put his UI skills to work to build Toad. Thenewly publicly released tool aims to provide a unified, "beautiful" GUI for multiple coding agents in your terminal, accessible via the same tool via the Agent Communication Protocol (ACP).
McGugan, who dedicated his last couple of years to building frameworks that make terminal applications more usable and appealing to users, believes that applications from companies in the AI space are less appealing because the technology stacks they use lack the appropriate building blocks.
/filters:no_upscale()/news/2025/12/llm-agent-cli/en/resources/12image3-1766158609046.png)
Image Source: batrachian.ai
The tool provides a front end for AI tools such as OpenHands, Claude Code, Gemini CLI, and other ACP-enabled agents, allowing them to run seamlessly within a single terminal application rather than separate CLIs. At the time of writing, it supports 12 agent CLIs and uses the ACP protocol to standardise communication, allowing new tools to be added as they appear. According to McGugan, he focused on making the onboarding as seamless as possible - if you have the native tool for a given LLM agent already set up, it’s enough to get started in Toad.
You can install it directly:
curl -fsSL batrachian.ai/install | sh
or via UV:
uv tool install -U batrachian-toad --python 3.14
Toad focuses on UX features that make terminals feel more like graphical coding assistants, including an "@" convention to pull project files into context via fuzzy search that respects .gitignore. The prompt editor supports keyboard and mouse navigation, selection, cut/copy/paste, and live Markdown syntax highlighting, even for code blocks before the closing fence is typed.
Image Source: batrachian.ai
It also streams Markdown responses efficiently, remaining responsive for large outputs while rendering tables and syntax-highlighted code fences instead of falling back to plain text. This approach aims to improve on many existing terminal AI tools that either do not render Markdown or only support a minimal subset.
Image Source: batrachian.ai
Shell integration is a core design goal: users can run full-colour, interactive TUI and regular CLI commands inline using a "!" prefix or preconfigured patterns that automatically trigger shell mode. Toad borrows tab completion semantics from typical shells, so commands and paths can be completed and cycled through with Tab, keeping interaction aligned with existing terminal habits.
Drawing inspiration from Jupyter, Toad lets users move through previous conversation blocks, reuse them, copy them to the clipboard or back into the prompt, and export content as SVG, with plans to deepen these notebook-like interactions. Installation instructions and further details are available on batrachian.ai and in the Toad repository. Although some features and UI polish are still needed, McGugan considers that the tool is already positioned as a daily driver for AI-assisted coding.
The tool is released as open-source under the AGPL 3.0 license, and the creator states in its release post that he intends to continue developing the tool full-time if he can attract sufficient user support through sponsorships.
About the Author
Olimpiu Pop
Show moreShow less