Toad
A terminal interface that brings different coding agents into one window alongside a real shell
Medium risk
We rate an entry medium when the tool runs code, makes network calls or reads project files. Check what exactly it does before installing.
Why this level
- Gives connected agents access to a real shell with persisted state
Install
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/toadDetects the agents on your machine, checks the risk and pins the version.
Other ways to install
Install the tool
pipx install batrachian-toadWithout pipx, pip install batrachian-toad works too.
Install Toad: curl -fsSL batrachian.ai/install | sh, then run it with toad and pick an agent from the list.
Other ways from the author
curl -fsSL batrachian.ai/install | shThis is third-party code. Review the repository files before installing.
What it does
Toad is a TUI application that launches and switches between coding agents such as Claude, Gemini, Codex and others inside one terminal window. Besides chatting with an agent, Toad provides a real shell: you can change directories and set environment variables, and that state persists between commands, which most agent interfaces don't support. The list of supported agents grows through the open Agent Client Protocol, so developers can add their own agent.
Who it is for. For developers who keep several coding agents around and want one terminal interface instead of switching between separate CLIs.
Good fit when
- You need to switch between Claude, Gemini, Codex and other agents without changing windows
- You need a shell that remembers directory changes and environment variables between agent commands
- You want to try a new agent through one interface without learning its own CLI
Not a fit when
- You work with a single agent through its own CLI and don't need to switch
- You need a graphical environment rather than a terminal interface
Example request
Launch Claude in Toad and show me what other agents I can connectLimitations
Licensed under AGPL-3.0, with a separate commercial license available where AGPL doesn't fit. The agents themselves still require their own install and authentication; Toad only provides a shared interface for them.
How to disable. Remove the toad binary installed by the script from the directory it was placed in (typically your user PATH).
Security check
- Gives connected agents access to a real shell with persisted state
README in short
The README presents Toad as a single terminal interface for AI agents like Claude, Gemini and Codex, blending a traditional shell workflow with agentic work. It stresses that Toad Shell, unlike most agent interfaces, actually keeps directory and environment-variable state. Install is a single command via an install script, and the list of supported agents grows through third-party additions via the Agent Client Protocol.
FAQ
How is Toad different from running an agent directly?
Toad adds one interface on top of different agents and a real shell with state (directory, environment variables) that persists between commands.
Can I add an agent that isn't on the list?
Yes, through the Agent Client Protocol: developers can wire their own agent into Toad.
Related
A skills library that gives coding agents a development process: brainstorming, planning, TDD, subagents and code review
Skills for real engineers by Matt Pocock
Skills For Real Engineers
Small composable skills for engineering with agents: plan grilling, TDD, bug diagnosis, code review and architecture
GitHub toolkit for spec-driven development: the specify CLI adds agent commands and skills to a project, from principles to implementation
Reference MCP servers
Model Context Protocol servers
Official reference MCP servers: Filesystem, Fetch, Git, Memory, Sequential Thinking, Time and Everything