Interpretable Context Methodology: Folder Structure as Agentic Architecture

TL;DR

This paper introduces Model Workspace Protocol (MWP), a filesystem-based method for orchestrating AI workflows that simplifies multi-step processes by replacing complex frameworks with folder structures and files.

Contribution

The paper proposes MWP, a novel filesystem-based approach for AI agent orchestration that reduces engineering overhead in sequential workflows.

Findings

Simplifies AI workflow management using folder structures.

Reduces engineering overhead compared to multi-agent frameworks.

Open source implementation available under MIT license.

Abstract

Current approaches to AI agent orchestration typically involve building multi-agent frameworks that manage context passing, memory, error handling, and step coordination through code. These frameworks work well for complex, concurrent systems. But for sequential workflows where a human reviews output at each step, they introduce engineering overhead that the problem does not require. This paper presents Model Workspace Protocol (MWP), a method that replaces framework-level orchestration with filesystem structure. Numbered folders represent stages. Plain markdown files carry the prompts and context that tell a single AI agent what role to play at each step. Local scripts handle the mechanical work that does not need AI at all. The result is a system where one agent, reading the right files at the right moment, does the work that would otherwise require a multi-agent framework. This approach applies ideas from Unix pipeline design, modular decomposition, multi-pass compilation, and literate programming to the specific problem of structuring context for AI agents. The protocol is open source under the MIT license.