An AI-powered shell that understands natural language and macros
Project description
Cliara - AI-Powered Shell
An intelligent shell wrapper that understands natural language and macros.
Quick Start
# Install
pip install -e .
# Setup environment
cp .env.example .env
# Edit .env and add your OPENAI_API_KEY
# Run
cliara
Documentation
- Complete Guide - Full documentation
- Quick Start - Get started in 5 minutes
What is Cliara?
Cliara wraps your existing shell and adds:
- 🗣️ Natural language commands with
?prefix (Phase 2) - 📦 Powerful macro system (create, edit, delete, run)
- 🔍 Semantic history search:
? find when I fixed the loginor? what did I run to deploy - 🛡️ Safety checks for dangerous operations
- 💾 Save last command as macro instantly
- 🔄 Persistent command history with arrow-key recall across sessions
- ✏️ Rename macros without recreating them
- 📂 Proper
cdhandling (changes Cliara's own working directory) - 🚀 Normal commands work unchanged
Installation
# 1. Clone the repo
git clone https://github.com/yourusername/cliara.git
cd cliara
# 2. Install dependencies
pip install -e .
# 3. Setup environment
cp .env.example .env
# Edit .env and add your OPENAI_API_KEY
# 4. Run
cliara
Usage
# Start Cliara
cliara
# Normal commands work
cliara:proj > ls -la
cliara:proj > git status
# Natural language (requires OPENAI_API_KEY)
cliara:proj > ? kill process on port 3000
# Create a macro
cliara:proj > macro add test
> echo Step 1
> echo Step 2
>
# Edit an existing macro
cliara:proj > macro edit test
> echo Updated Step 1
> echo Updated Step 2
>
# Run it
cliara:proj > test
# Rename a macro
cliara:proj > macro rename old-name new-name
# Save last command
cliara:proj > echo "hello"
cliara:proj > macro save last as hello
# cd works correctly (changes Cliara's own directory)
cliara:proj > cd src
cliara:src >
Features
Phase 1 ✅ (Complete)
- Shell wrapper with pass-through
- Interactive macro system (add, edit, delete, rename, show, run, search)
- Save last command as macro
- Persistent command history (
~/.cliara/history.txt) with arrow-key recall - Multi-tier safety checks
- Auto-configuration
Phase 2 ✅ (Complete)
- LLM integration (OpenAI)
- Natural language → commands
- Context-aware suggestions
- Semantic history search:
? find .../? when did I .../? what did I run ...— search past commands by intent (stored in~/.cliara/semantic_history.json)
Storage Backends
- JSON (default) - Simple file-based storage
- PostgreSQL - Scalable database backend for millions of macros
Requirements
- Python 3.8+
- Windows, macOS, or Linux
Windows Users
After installation, you may need to add Python Scripts to your PATH:
C:\Users\<YourName>\AppData\Local\Packages\PythonSoftwareFoundation.Python.3.12_qbz5n2kfra8p0\LocalCache\local-packages\Python312\Scripts
Or restart your computer for PATH changes to take effect.
Version
v0.2.0 - Phase 2 Complete
- ✅ Natural language support
- ✅ Windows compatibility fixes
- ✅ PostgreSQL backend support
License
MIT
See docs/README.md for complete documentation.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file cliara-0.2.0.tar.gz.
File metadata
- Download URL: cliara-0.2.0.tar.gz
- Upload date:
- Size: 115.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
957bc685245dfc5309bd669bad0bfb250e9ce2bc9d277b47f916c5e35f41fa82
|
|
| MD5 |
006b0f4da9bd3aac37aa0df6e655f793
|
|
| BLAKE2b-256 |
0fdd7836d320ea3862c1de7e6f624e68a9e2472df019b6e9a736b852b67a3b23
|
File details
Details for the file cliara-0.2.0-py3-none-any.whl.
File metadata
- Download URL: cliara-0.2.0-py3-none-any.whl
- Upload date:
- Size: 121.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.12.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
aeb516d3522155046d9cdffa8cc11799ee9f21f4e9db1a72910145bab36b0d83
|
|
| MD5 |
9387d8edbe6b52a92e4c5cd420a19a67
|
|
| BLAKE2b-256 |
ed3159395168134f30469bcb11704b7e550522495dee29976c2ffe21cbf7191e
|