Multi-Agent System provides a lightweight yet powerful toolkit for designing and executing multi-agent simulations. Users can create custom Agent classes to encapsulate decision-making logic, define Environment objects to represent world states and rules, and configure a Simulation engine to orchestrate interactions. The framework supports modular components for logging, metrics collection, and basic visualization to analyze agent behaviors in cooperative or adversarial settings. It’s suitable for rapid prototyping of swarm robotics, resource allocation, and decentralized control experiments.

PPT Agents is an open-source Python toolkit that converts PowerPoint presentations into AI agents. It parses PPTX files via python-pptx, extracts slide texts and metadata, and indexes them in a vector store using LangChain. With integrated OpenAI models, it enables users to interactively query slide content, generate concise summaries, and perform batch slide editing. Users can update text, add or remove slides based on prompts, and export modified PPT files seamlessly. This allows rapid insights and automated adjustments to presentations without manual slide-by-slide work.