Sawdust: A New Skeuomorphic Carpentry Simulator Integrating AI Agents and Model Context Protocol for DIY Woodworking

Key Takeaways

• Skeuomorphic Design Philosophy: Sawdust moves away from traditional CAD 'rectangle extrusion' in favor of a virtual shop environment featuring realistic tools like miter saws and router tables.
• AI Agent Integration: The platform utilizes the Model Context Protocol (MCP), allowing AI agents to perform complex parametric procedures and interact with the software on parity with human users.
• YAML-Based Operations: Every action within the simulator is defined via YAML, facilitating seamless communication between human interfaces and AI agents.
• Comprehensive Output Suite: Users can generate 3D printing files, STEP files, video walkthroughs, and detailed cut plans with life-size AR viewing capabilities.
• Autonomous Agent Feedback: AI agents are capable of identifying missing functionality and filing feature requests when they encounter limitations in the available toolset.

In-Depth Analysis

A Departure from Traditional CAD: The Skeuomorphic Approach

Sawdust introduces a specialized environment for woodworkers that prioritizes realism over abstract geometric modeling. While popular tools like Sketchup often rely on extruding shapes to create structures, Sawdust focuses on a skeuomorphic simulation of a physical workshop. This environment includes specific virtual machinery such as miter saws, table saws, and router tables, all operating according to real-world wood specifications. By simulating the actual physical constraints and operations of woodworking, the software provides a more intuitive transition from digital design to the physical workbench. This approach is particularly beneficial for users who find traditional CAD workflows disconnected from the tactile reality of carpentry.

Human-Agent Parity via Model Context Protocol (MCP)

One of the most technically significant aspects of Sawdust is its implementation of the Model Context Protocol (MCP). This protocol allows AI agents to operate within the simulator with the same level of access and capability as a human user. The system is built on a foundation of YAML-based operations, which serves as a common language for both species of users. Because every procedure is defined in a structured data format, AI agents can quickly parse, execute, and even create new parametric procedures.

This creates what the developer describes as a 'flywheel' effect: as agents develop procedures for tasks like stick framing, these procedures become simple function calls that subsequent agents or humans can utilize. Furthermore, the parity extends to the user interface, where human-readable procedures and agent-executable code are synchronized. This collaborative environment allows a user to send a simple sketch or photo to an agent and command it to 'build this with Sawdust,' effectively delegating the technical modeling and planning to the AI.

From Digital Vision to Physical Reality

Sawdust is designed to be more than just a visualizer; it is a comprehensive planning tool for physical construction. The software addresses the 'last mile' of DIY projects by providing a suite of export options that translate a digital model into actionable data. Users can generate a Bill of Materials (BOM) and a detailed cut plan, which are essential for purchasing lumber and executing precise cuts.

The inclusion of life-size Augmented Reality (AR) viewing allows woodworkers to see their projects in their intended physical space before a single piece of wood is cut. Additionally, the platform features a Notion-like interface for authoring build guides, ensuring that the knowledge generated during the design phase is preserved for the assembly phase. This end-to-end workflow aims to provide a high-quality alternative to mass-produced furniture, allowing users to create custom designs with the precision of professional-grade software.

Industry Impact

The emergence of Sawdust highlights a growing trend in the intersection of AI agents and specialized manual trades. By integrating MCP, the project demonstrates how niche industries—like custom carpentry—can leverage AI to handle complex, repetitive, or highly technical planning tasks. The ability for agents to file their own feature requests suggests a future where software evolves based on the needs of both human and artificial users. Furthermore, the shift toward skeuomorphic, tool-based simulation over abstract modeling could influence the next generation of CAD software, making professional-grade design more accessible to hobbyists and independent makers. This project underscores the potential for AI to act not just as a chatbot, but as a functional partner in physical manufacturing and DIY culture.

Frequently Asked Questions

Question: How does Sawdust differ from standard 3D modeling software like Sketchup?

Sawdust eschews the 'extruding rectangles' method of traditional CAD. Instead, it uses a skeuomorphic approach where users work within a virtual shop containing realistic tools (miter saws, router tables) and real wood specifications. It is built specifically for the logic of carpentry rather than general 3D geometry.

Question: What is the role of the Agent Model Context Protocol (MCP) in this tool?

The MCP allows AI agents to interact with Sawdust with full parity to human users. Agents can read and write YAML-based procedures, perform parametric tasks like stick framing, and even file feature requests for new tools they need to complete a project.

Question: What kind of files and data can Sawdust export for a physical build?

Sawdust can generate a variety of outputs including 3D printing files, STEP files, video walkthroughs, a Bill of Materials (BOM), and a cut plan. It also supports viewing projects at life-size scale using Augmented Reality (AR).