AI Assistant Authentication CLI

A powerful CLI tool for authenticating with Ollama Cloud and managing AI Assistant configurations. This tool replaces Google Cloud authentication with Ollama Cloud authentication, allowing you to use your self-hosted Ollama agent as a Google-search replacement.

Features

• 🔐 Ollama Cloud OAuth Authentication - Secure browser-based authentication flow
• ⚙️ Configuration Management - Easy setup and management of AI assistant settings
• 🌐 Self-hosted Integration - Connect to your Ollama instance running on OCI
• 📋 Status Monitoring - Check authentication and configuration status
• 🛠️ Global CLI Access - Install globally and use with npx ai-assistant-auth

Installation

Global Installation (Recommended)

npm install -g ai-assistant-auth

Local Installation

npm install ai-assistant-auth

Usage

Initialize Configuration

ai-auth init

Sets up the initial configuration file at ~/.ai-assistant/config.json

Authenticate with Ollama Cloud

ai-auth auth

Opens your browser to authenticate with Ollama Cloud and obtain your API key.

Check Status

ai-auth status

Displays current authentication status and configuration details.

Manage Configuration

# List all configuration settings
ai-auth config --list

# Get a specific configuration value
ai-auth config --get "assistant.defaultModel"

# Set a configuration value
ai-auth config --set "assistant.defaultModel=llama3.2"

Using Your Authenticated Ollama API

After authentication, you can use your OLLAMA_API_KEY with the Ollama client:

Node.js Example

import { Ollama } from "ollama";

const ollama = new Ollama({
  host: "https://ollama.com",
  headers: {
    Authorization: "Bearer " + process.env.OLLAMA_API_KEY,
  },
});

const response = await ollama.chat({
  model: "llama3.2",
  messages: [{ role: "user", content: "Explain quantum computing" }],
  stream: true,
});

for await (const part of response) {
  process.stdout.write(part.message.content);
}

Python Example

import os
from ollama import Client

client = Client(
    host="https://ollama.com",
    headers={'Authorization': 'Bearer ' + os.environ.get('OLLAMA_API_KEY')}
)

messages = [
  {
    'role': 'user',
    'content': 'Why is the sky blue?',
  },
]

for part in client.chat('llama3.2', messages=messages, stream=True):
  print(part['message']['content'], end='', flush=True)

Environment Variables

After authentication, set your OLLAMA_API_KEY:

export OLLAMA_API_KEY="your-api-key-from-authentication"

Commands

| Command | Description | |---------|-------------| | ai-auth auth | Authenticate with Ollama Cloud | | ai-auth init | Initialize configuration | | ai-auth status | Check authentication status | | ai-auth config --list | List all configuration | | ai-auth config --get <key> | Get configuration value | | ai-auth config --set <key=value> | Set configuration value |

Configuration File

The tool creates a configuration file at ~/.ai-assistant/config.json with default settings:

{
  "ollama": {
    "baseUrl": "http://localhost:11434",
    "cloudEnabled": true
  },
  "assistant": {
    "defaultModel": "llama3.1",
    "temperature": 0.7,
    "maxTokens": 2048
  },
  "features": {
    "chat": true,
    "fileUpload": true,
    "camera": true
  }
}

Development

Building from Source

npm run build

Running Development Version

npm run dev

Testing CLI

node dist/cli.js --help

Publishing to npm

To publish updates:

npm login
npm publish

License

MIT

Author

AI Assistant Deployment Hub