AdaL CLI
The AdaL terminal agent: switches models, shows edits and plans for review and hands the session over to its own IDE.
- 111 ★ on the repository
- Last commit
Broad access
Why this level
- The agent runs shell commands, changes project files and reaches the network.
- It needs browser authentication and a service account.
Install
Manual install
curl -fsSL https://adal.sylph.ai/install.sh | bashmacOS, Linux and WSL; then run `adal`.
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- The agent runs shell commands, changes project files and reaches the network.
- It needs browser authentication and a service account.
This is third-party code. Review the repository files before installing.
What it does
AdaL CLI is a terminal agent harness that orchestrates models, tools and agents into end-to-end workflows from issue to tested pull request. It was made by the SylphAI team. It works with Claude, GPT, Gemini, GLM, Kimi, DeepSeek, MiniMax and local models. Tool calls, file edits, diffs and plans stay visible until you approve them. The `/model` command switches the model, `/ide` opens the current session in the AdaL IDE, `/resume` returns to earlier work with context and `/stats` shows session health and usage. According to the docs, the agent supports skills and plugins, MCP server connections, deep research, web search, real browser use, image generation and analysis, headless mode, scheduled prompts, local models and your own API keys. The repository is the community home: discussions, questions and hackathon material.
Who it is for. Developers looking for a terminal agent that switches among many models and can move into a graphical environment.
Good fit when
- You want to compare how Claude, Gemini and DeepSeek solve one task by switching with `/model`.
- A session should continue in an IDE where files, terminal and canvas are available together.
- A task must resume through `/resume` with saved context.
- You need to run the agent in a script or on a schedule without a UI.
Not a fit when
- You need fully open agent source: the repository holds no CLI code.
- You cannot sign in through a browser: the first run opens an authentication page.
- You need an agent that works strictly offline with no account.
Example request
Take the issue from the tracker, propose a plan, show it for approval, then make the edits and open a pull request.Limitations
The repository holds community materials only: there is no CLI source code, and installation comes from the developer's site. The first run opens a browser for authentication. Supported models, pricing and plans are listed in the developer docs, and no license terms are declared in the repository. The agent runs commands and changes files, so keep tool confirmations on. Working with models needs a service account or your own API keys.
How to disable. Remove the installed `adal` program and its settings folder. The procedure for your system is described in the developer docs.
FAQ
How do I install AdaL CLI?
On macOS, Linux and WSL with `curl -fsSL https://adal.sylph.ai/install.sh | bash`, on Windows with `irm https://adal.sylph.ai/install/windows | iex`. Then run `adal`.
Which models are supported?
Claude, GPT, Gemini, GLM, Kimi, DeepSeek, MiniMax and local models. You can also bring your own API keys.
Which commands matter most?
`/model` picks the model, `/ide` opens the session in the IDE, `/resume` returns to earlier work and `/stats` shows session health and usage.
What is in this repository?
Community materials: banners, the `@skills:` hackathon presentation and links to docs. The CLI code is not published here.
Where is the full documentation?
At docs.sylph.ai: quickstart, models and billing, shortcuts, skills and plugins, MCP, browser use, headless mode and scheduled prompts.
Discussion
No questions or reviews yet.
Related
CLI toolsAn open coding agent for the terminal, browser and desktop that works with models from OpenAI and Claude to Ollama
Agav is a terminal coding agent that reads a repository, edits files, runs tests and switches between seven model providers.
A terminal coding agent in one Rust binary that reads code, edits files and runs tests with any model of your choice.
A Go CLI that configures the AI coding agents you already use: it gives them shared memory, a workflow and deterministic steps without replacing the agents them
A local control panel for coding agents: it runs Claude Code and Codex, keeps a task queue and merges verified work into main.
A local-first AI agent running on your own machine via llama.cpp that drives a browser, edits files, runs approved commands and remembers context across session
README in short
The repository is the community around the AdaL product: it holds discussions and materials. The agent itself installs from the developer site.