Skip to main content

TerminalAI: Command-line AI assistant

Project description

Terminal AI

Bring the power of AI directly to your command line!

TerminalAI is your intelligent command-line assistant. Ask questions in natural language, get shell command suggestions, and execute them safely and interactively. It streamlines your workflow by translating your requests into actionable commands.

████████╗███████╗██████╗ ███╗   ███╗██╗███╗   ██╗ █████╗ ██║       █████╗ ██╗
╚══██╔══╝██╔════╝██╔══██╗████╗ ████║██║████╗  ██║██╔══██╗██║      ██╔══██╗██║
   ██║   █████╗  ██████╔╝██╔████╔██║██║██╔██╗ ██║███████║██║      ███████║██║
   ██║   ██╔══╝  ██╔══██╗██║╚██╔╝██║██║██║╚██╗██║██╔══██║██║      ██╔══██║██║
   ██║   ███████╗██║  ██║██║ ╚═╝ ██║██║██║ ╚████║██║  ██║███████╗ ██║  ██║██║
   ╚═╝   ╚══════╝╚═╝  ╚═╝╚═╝     ╚═╝╚═╝╚═╝  ╚═══╝╚═╝  ╚═╝╚══════╝ ╚═╝  ╚═╝╚═╝

Key Features

  • Natural Language Interaction: Ask questions or request actions naturally.
  • Intelligent Command Suggestion: Get relevant shell commands based on your query.
  • File Reading & Explanation:
    • Use --read-file <filepath> along with your query to have the AI consider a file's content.
    • Use --explain <filepath> for a direct summary and contextual explanation of a file.
    • Supports various plain text files; the AI attempts to interpret the content.
  • Multiple AI Backends: Supports OpenRouter, Gemini, Mistral, and local Ollama models.
  • Interactive Execution: Review and confirm commands before they run.
  • Context-Aware: Includes OS and current directory information in prompts to the AI.
  • Safe Command Handling:
    • Non-stateful commands run directly after confirmation.
    • Risky commands require explicit confirmation.
    • Stateful commands (cd, export, etc.) are handled safely (see below).
  • Multiple Modes:
    • Direct Query (ai "..."): Get a single response and command suggestions.
    • Single Interaction (ai): Ask one question, get a response, and return to the shell.
    • Chat Mode (ai --chat or ai -c): Persistent conversation with the AI.
  • Easy Configuration: ai setup provides a menu for API keys and settings.
  • Optional Shell Integration: For seamless execution of stateful commands in direct query mode.
  • Syntax Highlighting: Uses rich for formatted output.

Installation

Option 1: Install from PyPI (Recommended)

pip install coaxial-terminal-ai

Option 2: Install from Source

git clone https://github.com/coaxialdolor/terminalai.git
cd terminalai
pip install -e .

This automatically adds the ai command to your PATH.

Quick Setup

  1. Install: Use one of the methods above.
  2. Configure API Keys: Run ai setup and select option 5 to add API keys for your chosen provider(s) (e.g., Mistral, Ollama, OpenRouter, Gemini).
  3. Set Default Provider: In ai setup, select option 1 to choose which provider ai uses by default.
  4. (Optional) Install Shell Integration: See "Handling Stateful Commands" below if you want direct execution for commands like cd when using ai "...".
  5. Start Using: You're ready!

See the Quick Setup Guide for more detailed instructions.

Usage Examples

1. Single Interaction Mode (ai): Ask one question, get an answer/commands, then return to shell. Flags like -v or -l can be used here.

# Basic usage
ai
AI:(mistral)> how do I list files by size?

# Request a long response
ai -l
AI:(mistral)> explain the history of Unix shells in detail

2. Direct Query Mode (ai "..."): Provide the query directly. This is where most flags are useful.

# Simple query
ai "find all python files modified in the last day"

# Auto-confirm non-risky command execution
ai -y "show current disk usage"
# (Example: If AI suggests 'df -h', it will run without a [Y/n] prompt)

# Request verbose output
ai -v "explain the concept of inodes"

# Request long output
ai -l "explain the difference between TCP and UDP"

# Combine flags: Auto-confirm and Verbose
ai -y -v "create a new directory called 'test_project' and list its contents"
# (Example: If AI suggests 'mkdir test_project && ls test_project', it will run without a prompt)

# Read and explain a file
ai --read-file ./my_script.py "Summarize this Python script and what it does"

# Get an automatic explanation of a file
ai --explain ./config/app_settings.yaml

3. Chat Mode (ai --chat or ai -c): Have a persistent conversation.

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

coaxial_terminal_ai-0.3.0.tar.gz (45.4 kB view details)

Uploaded Source

Built Distribution

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

coaxial_terminal_ai-0.3.0-py3-none-any.whl (49.6 kB view details)

Uploaded Python 3

File details

Details for the file coaxial_terminal_ai-0.3.0.tar.gz.

File metadata

  • Download URL: coaxial_terminal_ai-0.3.0.tar.gz
  • Upload date:
  • Size: 45.4 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.1.0 CPython/3.13.0

File hashes

Hashes for coaxial_terminal_ai-0.3.0.tar.gz
Algorithm Hash digest
SHA256 cbfb1c6fe8238efd70bf504014c7f85a28bb713501fc8b5d249e65ad3793e344
MD5 6f03674f9d9fd0bb25730ecaf45fad07
BLAKE2b-256 a24a44e8d2a3d016014bfb60191d312f6987fc6b96750e072b45b0750513366e

See more details on using hashes here.

File details

Details for the file coaxial_terminal_ai-0.3.0-py3-none-any.whl.

File metadata

File hashes

Hashes for coaxial_terminal_ai-0.3.0-py3-none-any.whl
Algorithm Hash digest
SHA256 19a81dcbcf01a6cf5f62d552b01e949f8179c7e8cea58ee41fda50b26f4aa1fb
MD5 f14286ba3ee75849a7d6eb390558aad8
BLAKE2b-256 ce582e3fb87cd0e0b2f8a2d4156c300080e01d2e942e7a361bc9d85b72351f2b

See more details on using hashes here.

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