3.0.15 • Published 6d agoCLI

@atonis/cli

Licence

Apache-2.0

Version

3.0.15

Deps

Size

79.1 MB

Vulns

Weekly

Summary Dependency Versions

Atonis CLI

NPM

Website

Docs

Issues

Meet Atonis — an autonomous AI coding assistant that lives in your terminal, brought to you by Ness Digital Engineering.

Atonis can handle complex software development tasks step-by-step. With tools that let it create & edit files, explore large projects, use the browser, and execute terminal commands (after you grant permission), it can assist you in ways that go beyond code completion or tech support.

npm i -g atonis

# cd into your project and run:
atonis

Features

Use Any API and Model

Atonis supports API providers like Anthropic, OpenAI, Google Gemini, AWS Bedrock, Azure, GCP Vertex, Cerebras, Groq, OpenRouter, Ollama, and more. You can also configure any OpenAI-compatible API, or use a local model through LM Studio/Ollama.

Stay in Control with Human-in-the-Loop

Atonis asks for your approval before running commands, editing files, or taking any action. Review each step and approve or reject as you go — or enable auto-approve to let Atonis work autonomously to completion.

Plan & Act Modes

Toggle to Plan Mode to discuss implementation and architecture with Atonis. It will ask clarifying questions, explore your codebase, and present a plan for you to align on. Once you're satisfied, switch to Act Mode and let Atonis execute the plan.

Installation

npm install -g atonis

Requires Node.js 20+.

Quick Start

# Launch interactive mode
atonis

# Run a task directly
atonis "Create a hello world function in Python"

# Run in yolo mode (auto-approve all actions)
atonis -y "Fix the typo in README.md"

# Run in plan mode first
atonis -p "Design a REST API for user management"

Authentication

# Interactive authentication wizard
atonis auth

# Quick setup for Anthropic
atonis auth -p anthropic -k sk-ant-api-xxxxx

# Quick setup for OpenAI
atonis auth -p openai-native -k sk-xxxxx -m gpt-4o

# OpenAI-compatible provider with custom base URL
atonis auth -p openai -k your-api-key -b https://api.example.com/v1

Commands

Command	Description
`atonis [prompt]`	Start a task (or open interactive mode)
`atonis task [prompt]`	Alias for running a task
`atonis history`	List previous tasks
`atonis auth`	Configure API provider and model
`atonis config`	Show current configuration
`atonis update`	Check for and install updates
`atonis version`	Show the CLI version

Task options

Flag	Description
`-a, --act`	Run in act mode (default)
`-p, --plan`	Run in plan mode
`-y, --yolo`	Auto-approve all actions; forces plain text output
`-m, --model <id>`	Model to use for the task
`-t, --timeout <seconds>`	Task timeout in seconds
`-i, --images <paths...>`	Image file paths to include
`-v, --verbose`	Show verbose output including reasoning
`-c, --cwd <path>`	Working directory for the task
`--thinking`	Enable extended thinking (1024 token budget)
`--json`	Output messages as JSON
`-T, --taskId <id>`	Resume an existing task by ID
`--continue`	Resume the most recent task in the current directory

Scripting & Automation

# Pipe file contents
cat README.md | atonis "Summarize this document"

# JSON output for parsing
atonis --json "What files are in this directory?" | jq '.text'

# Yolo mode for automated workflows
atonis -y "Run the test suite and fix any failures"

# Resume a previous task
atonis --continue "Now add unit tests for the changes"

Configuration Files

~/.atonis/
└── data/
    ├── globalState.json     # Global settings and state
    ├── secrets.json         # API keys (stored securely, mode 0o600)
    ├── tasks/               # Task history and conversation data
    └── workspaces/          # Per-workspace state