Skip to main content

Command-line interface for Navari AI Agent

Project description

Navari CLI

Command-line interface for Navari AI Agent - an enterprise-grade AI agent framework for pharmaceutical research.

Installation

uv tool install navari-cli
# or
pip install navari-cli

# Upgrade to latest version
uv tool upgrade navari-cli
# or
pip install --upgrade navari-cli

Quick Start

# Login to Navari
navari login https://navari.example.com

# List conversations
navari conversations list

# Start interactive chat
navari chat --new --title "My Analysis"

# Get help
navari --help

Features

  • Interactive Chat: Real-time streaming chat with Navari agent
  • Conversation Management: Create, list, get, update, delete conversations
  • File Operations: Upload, download, and manage files
  • Multiple Servers: Switch between servers with navari config <url>
  • Azure AD SSO: Browser-based, device code, and password authentication flows
  • Beautiful Output: Rich terminal output with colors and tables

Commands

Authentication

navari login <url>                         # Browser-based OAuth flow
navari login <url> --device-code           # Device code flow for headless
navari login <url> --token <jwt>           # Login with a pre-obtained JWT token
navari login <url> -u user@co.com -p pass  # Password login (CI/automation)
navari login <url> --no-verify             # Skip SSL verification (remembered)
navari login                               # Re-login to last used server
navari logout                              # Logout from all servers
navari logout <url>                        # Logout from specific server
navari whoami                              # Show current user and server

Conversations

navari conversations list                   # List all conversations
navari conversations create --title "Title" # Create new conversation
navari conversations get <id>               # Get conversation details
navari conversations messages <id>          # Show conversation messages
navari conversations delete <id>            # Delete conversation

Interactive Chat

navari chat <conversation-id>               # Start chat session
navari chat --new --title "Title"           # Create new and start chat
navari chat -m "message"                    # Send single message (non-interactive)
navari chat --new -f data.csv -m "analyze"  # Attach file and send message

Files

navari files upload --conversation <id> <path>  # Upload file
navari files list --conversation <id>           # List files
navari files download <file-id>                 # Download file
navari files delete <file-id>                   # Delete file
navari files workspace <conversation-id>        # Download workspace zip

Configuration

navari config                               # Show current configuration
navari config <url>                         # Switch active server

Global Options

--server <url>               # Override active server for this command
--format {json|table|plain}  # Output format
--debug                      # Enable debug logging
--no-color                   # Disable color output

Configuration

The CLI stores credentials and configuration in ~/.navari/:

  • credentials.json - JWT tokens per server
  • config.json - User preferences (last used server, output format)
  • msal_cache.json - MSAL token cache for silent token refresh

Development

Generating OpenAPI Client

When the Navari API changes, regenerate the client:

./scripts/generate_api.sh --client-py

See Generated Client for details on the generated API client.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

navari_cli-0.5.3.tar.gz (93.3 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

navari_cli-0.5.3-py3-none-any.whl (196.4 kB view details)

Uploaded Python 3

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page