Toad is an open-source terminal user interface (TUI) coding agent harness developed by Batrachian AI and spearheaded by Will McGugan (creator of Rich and Textual). Built natively in Python, Toad provides a structured terminal environment for orchestrating AI-driven code edits, tool execution, and interactive developer workflows. In daily use, Toad operates within the terminal, offering an interactive UI to manage AI agent execution. It incorporates prompt interaction, model configuration (including support for providers such as Moonshot/Kimi and standard LLM APIs), and an integrated permission flow that allows developers to review, approve, or deny agent actions and file modifications before execution. What distinguishes Toad from standard script-based coding CLI tools is its rich TUI design and explicit UI-driven permission handling. Instead of executing unchecked shell operations or dumping raw streams to stdout, Toad organizes agent thoughts, tool requests, and permission prompts into dedicated terminal panels. The project is licensed under open-source terms with dedicated commercial licensing provisions and active ongoing maintenance across the repository.
Terminal User Interface (TUI) — Provides an interactive terminal layout for monitoring agent decisions and reviewing diffs. - Interactive Permission System — Presents explicit permission prompts and question views before executing sensitive tools or writing files. - Multi-Model Provider Support — Connects to major LLM backends and specialized providers such as Kimi for reasoning and code generation. - Python-Native Architecture — Implemented in modern Python with structured package dependencies and type safety. - Commercial and Open Source Licensing — Distributed with clear guidelines and dual-licensing paths for enterprise integration.
Use cases
Use Case: Interactive Terminal Code Generation Developers use Toad to execute complex codebase refactorings from within their existing terminal workflow. The harness renders multi-step code generation tasks while keeping diffs clearly visible.
Use Case: Safe Tool and Shell Execution Teams running autonomous or semi-autonomous tasks leverage Toad's built-in permissions view to validate every shell command and file edit before the agent touches disk.
Use Case: Provider-Flexible Agent Experimentation Engineers wanting to test emerging LLM providers (such as Kimi) can plug alternative model configurations directly into a dedicated terminal interface without switching client tools.
Who it is for
Toad is designed for terminal-centric software engineers, Python developers, and CLI power users who want a structured TUI for agentic coding rather than bare CLI prompts. It is less suited for developers seeking deep IDE GUI integration or zero-configuration graphical desktop applications.
Install guide
Installation Toad is distributed as a Python package. You can install it using pip, pipx, or directly from the repository: bash # Recommended installation via pipx pipx install toad # Or install from source via git git clone https://github.com/batrachianai/toad.git cd toad pip install -e .
First Run Set your desired LLM API key in your environment variables: bash export OPENAI_API_KEY="your-api-key-here" # Or export provider-specific keys (e.g., MOONSHOT_API_KEY / KIMI_API_KEY) Launch Toad in your project's root directory: bash toad
Example Session Start an interactive session to inspect and modify code: bash # Inside the Toad TUI prompt > Refactor src/toad/permissions.py to include max width constraints and run tests Toad will plan the task, request tool execution approvals through its permissions interface, and display real-time terminal output.
agent-browser — agent-browser is an open-source browser automation harness and agent tool developed by Vercel Labs. It provides autonomous AI agents with…
Codel — Codel is an open-source, self-hosted autonomous software engineering agent created by semanser. It combines an execution backend, an interactive web…
agent-trace — agent-trace is an open-source command-line transcript utility created by ertygiq. It is designed to inspect, parse, and export execution logs…
Agent Manager — Agent Manager is an open-source Go-based CLI utility and orchestration harness created by Kevin Elliott. It is designed to catalog,…
Amazon Q Developer CLI — Amazon Q Developer CLI is an open-source command-line coding assistant developed by Amazon Web Services (AWS). Built primarily in Rust,…
SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…