aiexec

Run natural-language .ai automation scripts with a local, safe, agentic CLI loop.

Quick Install

One-line bootstrap (macOS/Linux)

curl -fsSL https://raw.githubusercontent.com/your-org/aiexec/main/install/install-aiexec.sh | bash

One-line bootstrap (Windows PowerShell)

irm https://raw.githubusercontent.com/your-org/aiexec/main/install/install-aiexec.ps1 | iex

Usage

Run the provided example task:

aiexec run examples/setup_fastapi.ai --dry-run

• --yes auto-confirms dangerous commands (e.g., rm -rf); omit to require confirmation.
• --sandbox executes inside a temporary directory instead of your current working tree.
• --verbose prints raw model responses.
• --dry-run shows proposed commands without executing them.

Other helpful commands:

aiexec init my_task         # create my_task.ai template
aiexec list                 # list .ai files in the current directory
aiexec doctor               # check for Ollama/Hugging Face setup issues

Requirements

• Python 3.10+
• Ollama with the llama3:instruct model (ollama pull llama3:instruct)
• Optional: HUGGINGFACE_API_TOKEN environment variable for Hugging Face models (--model hf:<repo>).

Development

Format and lint (optional):

pip install black ruff
black aiexec
ruff check aiexec

Run the unit-less sanity check:

python -m compileall aiexec