AutoGRAMS: Autonomous Graphical Agent Modeling Software

TL;DR

AutoGRAMS is a framework for designing interpretable, controllable, and safe AI agents using a graph-based structure that integrates language models and traditional code, enabling complex multi-step interactions and self-modification.

Contribution

AutoGRAMS introduces a novel graph-based framework for programming AI agents that combines language models with traditional logic, supporting self-referential and modular agent design.

Findings

Supports complex multi-step agent interactions

Enhances interpretability and safety in AI agent design

Enables self-modifying agent behaviors

Abstract

We introduce the AutoGRAMS framework for programming multi-step interactions with language models. AutoGRAMS represents AI agents as a graph, where each node can execute either a language modeling instruction or traditional code. Likewise, transitions in the graph can be governed by either language modeling decisions or traditional branch logic. AutoGRAMS supports using variables as memory and allows nodes to call other AutoGRAMS graphs as functions. We show how AutoGRAMS can be used to design highly sophisticated agents, including self-referential agents that can modify their own graph. AutoGRAMS's graph-centric approach aids interpretability, controllability, and safety during the design, development, and deployment of AI agents. We provide our framework as open source at https://github.com/autograms/autograms .