openmacro 0.1.7

Multimodal Assistant. Human Interface for computers.

Project is in its early stage of development.

openmacro is a multimodal personal agent that allows LLMs to run code locally. openmacro aims to act as a personal agent capable of completing and automating simple to complex tasks autonomously via self prompting.

This provides a cli natural-language interface for you to:

• Complete and automate simple to complex tasks.
• Analyse and plot data.
• Control desktop applications through vision and pyautogui.
• Manipulate files including photos, videos, PDFs, etc.

At the moment, API keys for models are yet to be supported and by default, models are powered by SambaNova with fallback powered by HuggingFace Spaces and their respective hosts:

• LLM: Llama3.1-405B-Instruct hosted by Corvius.
• Code: CodeQwen1.5-7B hosted by Qwen.
• Vision: Qwen-VL-Max hosted by Qwen.

This project is heavily inspired by Open Interpreter.

Quick Start

You can get started with openmacro by running.

pip install openmacro
macro

Not working? Raise an issue here or try this out instead:

py -m pip install openmacro
py -m openmacro

Personalisation

Openmacro supports cli args and customised settings! You can view arg options by running:

macro --help

To append your own personalised settings and save it for the future, run:

macro --config "path\to\config.toml" --save

What your personalised config.toml might look like:

[assistant]
name="Basil"
personality="You have a kind, deterministic and professional attitude towards your work and respond in a formal, yet casual manner."

Todo's

• AI Intepreter
• Web Search Capability
• Async Chunk Streaming
• Polished Extensions Feature
• Cost Efficient Long Term Memory & Context Manager
• API Keys Support
• Security & Error Handling
• Semantic File Search
• App Interface

Currently Working On

Currently, focusing on a refined system for Extensions, so anyone can build extensions for openmacro.

• Working on browser which is the official openmacro browser extension powered by Playwright, Bs4 and Markdownify.

• Working on ompi which is the Openmacro Package Index. Similar to Python's pypi, users can install third-party extensions to integrate with their openmacro assistant.

• Cost efficient long term memory and conversational context managers through vector databases. Most likely powered by Milvus or Milvus Lite.

Next Steps

Better integration of extensions.

Contact

You can contact me at [e-mail]