How to use Claude to build a terminal workspace
Bothy packages a Zellij and Yazi terminal workspace into a one-command setup. Claude wrote most of it under the author’s decisions and rules, with portable configuration and verified official tool downloads.
Prompt to try
Editorial adaptation, not the source author’s exact prompt. Replace bracketed inputs and review tool permissions first.
Build [portable terminal workspace] around [terminal multiplexer] and [file browser], following [project rules]. Plan and implement a single launch command, configure existing tools or verify official downloads, and keep installed files in [removable directory]. Let me decide scope and record rejected feature ideas.
Reach people exploring AI agents
See placements →What happened
Bothy opens a terminal workspace with one command, configures existing tools or downloads verified official releases, and keeps its installation in one removable directory.
Reported by bspeelm on Hacker NewsHow the agent did it
- Assembled a terminal workspace with Zellij, Yazi and a development console alias.
- Rebuilt the setup on a laptop and work machine, exposing portability differences.
- Set project decisions and rules and had Claude plan and execute Bothy within them.
Where a person stepped in
The author made decisions and set rules. The post does not describe specific code review or test steps.
A person approved each step before the agent continued.
The author wanted a development workspace centered on three tools: a file browser, a console for code assistance and a console for CLI commands. They began with Zellij and Yazi and a development console alias. Rebuilding the setup on a laptop and a work machine produced slightly different versions, prompting the portability project called Bothy.
The resulting tool opens a workspace with one command. If the required tools already exist, it sets up configuration and an alias. Otherwise, it downloads official releases, verifies them and keeps everything inside one directory that can be uninstalled. The author reports no root requirement or PATH changes, and says the workspace can use a user's own dotfiles rather than its defaults.
Claude wrote most of the project. The author deliberately used this low-stakes work to test an AI-led workflow: they made decisions and set rules, while Claude planned and executed within those boundaries. Documentation covers the experiment, including decisions against building some features, and the author identifies claude.md as its final state. No detailed test or code-review procedure appears in the supplied post. Who it's for: developers who want a portable terminal workspace and can define clear limits for agent-led implementation.
Prompt
Suggested prompt (written by Agents In Use, not the author's)
Source: Hacker News · bspeelm
Frequently asked questions
What does Bothy set up?
It opens a workspace based on Zellij and Yazi. Existing tools receive configuration and an alias; missing tools are downloaded from official releases and verified.
Who controlled the AI-led workflow?
The author says they made decisions and set rules, while Claude planned and executed inside those boundaries. Claude wrote most of the project.
More details
About this account
Claude planned and executed within author-defined rules and wrote most of the project; the author retained decisions.
Platform: Hacker News. Attribution: https://news.ycombinator.com/item?id=49553789. Prompt: Suggested (written by Agents In Use, not the author's). Sourced by the Agents In Use pipeline.
Credit: bspeelm on Hacker News
Source published 3 Sept 2026
Source checked 14 Oct 2026
Tools and topics
Claude · Zellij · Yazi · Bothy
terminal workspace · portability · developer tools
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.