Developer Brandon Thomas has released WordCraft, GridCraft, and DeckCraft, a suite of free, open-source clones of Microsoft Word, Excel, and PowerPoint. Built using the Rust programming language with assistance from Anthropic’s AI model Claude Opus 5.5, these applications are available on GitHub under MIT and Apache 2.0 licenses, allowing for commercial reuse. The release follows Thomas’ earlier ArtCraft project, which cloned Adobe Creative Cloud applications, and targets users paying $99.99 annually for Microsoft 365 Personal subscriptions.
The applications are described as clean-room rebuilds, developed by observing original software behavior and utilizing public file-format specifications rather than copying Microsoft’s proprietary code. This approach mirrors historical precedents such as Compaq’s cloning of IBM’s PC BIOS in the 1980s. WordCraft supports .docx files and covers 88% of Word’s ribbon features, including track changes and mail merge, though the team estimates it is only 60% ready for professional use. GridCraft handles 513 of Excel’s 523 worksheet functions and 89% of ribbon commands but is rated approximately 50% ready for real-world work due to compatibility issues with complex spreadsheets. DeckCraft covers about 79% of PowerPoint’s features and supports .pptx files, transitions, and PDF exports, but has not yet reached its first alpha release.
A distinctive feature of the suite is that every button acts as a command triggerable by an AI agent via the Model Context Protocol (MCP), enabling automated tasks such as chart creation or report formatting without manual input. While the apps offer significant cost savings and offline functionality, they remain in early testing stages. WordCraft has garnered over 12,000 GitHub stars, indicating strong community interest despite Thomas’ previous delays in meeting feature parity promises.
Brandon Thomas’ release of WordCraft, GridCraft, and DeckCraft demonstrates the accelerating capability of large language models to generate functional, cross-platform software alternatives to established enterprise suites. By leveraging Rust for stability and MCP for AI integration, the project highlights a shift where development speed and feature coverage are increasingly driven by AI-assisted coding rather than traditional human engineering timelines. However, the self-reported readiness metrics—ranging from 50% to 60%—underscore the persistent gap between syntactic correctness and semantic reliability in complex office workflows, particularly regarding file fidelity and edge-case handling.
From a market structure perspective, this development poses a credible threat to Microsoft’s subscription dominance by offering a zero-cost, offline alternative that mimics the user interface and core functionalities of Office. The strategic inclusion of native AI agent control via MCP differentiates these tools from legacy open-source competitors like LibreOffice, positioning them as platforms for automated productivity rather than just static document editors. Nevertheless, widespread institutional adoption remains hindered by the lack of comprehensive testing against real-world corporate data structures and the absence of formal support channels, limiting their immediate utility to individual power users and developers willing to tolerate alpha-stage instability.
