Sanitizer

The description states that Sanitizer is a Windows desktop application. It is designed to:

• Process individual files or complete folders.
• Extract text, images, metadata, and tables from supported documents.
• Apply user-defined rules for deletion or masking of sensitive content.
• Convert supported content into Markdown.
• Organize processed documents by type.
• Generate a searchable index.
• Export a folder-based workspace for Codex.

It is built using .NET, C#, and Windows App SDK technologies. The application preserves original files and operates locally on the user’s computer.

Inference The product is described as a local desktop tool, not a cloud or SaaS offering. It is intended to be used by managers in engineering environments who need to sanitize documents before AI processing.

The author states that Sanitizer addresses the concern that “privacy and sensitive information protection is the biggest concern when using AI.” The product is positioned as a solution for engineers and managers who work with confidential project documentation and want to safely prepare it for AI tools like Codex.

It claims to be:

• A simple, non-technical tool for document sanitization.
• A local-first solution, keeping data on the user’s machine.
• A bridge between real-world documents and AI-assisted workflows.
• A manager-friendly interface that abstracts complex processes like OCR, indexing, and verification.

Inference Sanitizer positions itself as a sanitization tool for AI-ready engineering documentation, not a general-purpose document editor or AI platform. It is framed as solving a trust and workflow problem in AI adoption.

The description states that Sanitizer is designed for managers working with engineering project packages, including:

• Specifications
• Control narratives
• Sequences of operation
• Building drawings
• Electrical drawings
• Mechanical drawings
• P&IDs (Piping and Instrumentation Diagrams)

These are described as containing sensitive data such as customer names, site information, addresses, and network details.

Inference The target customer is non-technical professionals in engineering or project management, who need to sanitize confidential documents before AI processing. The ICP appears to be engineering teams or project managers working with large, structured documentation sets.

Not evidenced.

The description does not mention any pricing model, monetization strategy, or business model. It is unclear whether Sanitizer will be sold, offered as a free tool, or distributed through a subscription.

The project is built as a native Windows desktop application using:

• .NET
• C#
• Windows App SDK (WinUI 3)
• Microsoft Store distribution
• OpenXML SDK
• OCR
• SQLite for local indexing
• Codex (GPT-5.6 Sol) for development assistance

It is described as a folder-and-file-based system, with no proprietary database dependency.

Inference The technical stack suggests a local-first, Windows-centric solution. The use of Microsoft Store distribution and WinUI 3 indicates an intent to reach users through official Windows channels.

Not evidenced.

There is no mention of:

• Revenue
• Customers
• Usage metrics
• Product adoption
• Market feedback or user testing

The project is described as a working desktop application, but no evidence of traction beyond the hackathon submission exists.

Not evidenced.

No information is provided about competitors, market size, or competitive positioning. The author does not reference existing tools for document sanitization or AI preparation in engineering workflows.

• No revenue or customer data: The product is described as a working prototype, but no evidence of monetization or adoption.
• Single-founder team: The project is built by one person (Mikhail Kozyrev), which may limit scalability or long-term development capacity.
• Limited scope: It is focused on Windows desktop and engineering documents. No indication of cross-platform support or broader applicability.
• No verification of claims: The description does not include any independent validation, testing, or user feedback.

1. What specific types of engineering or business documents are supported?
2. How is the sanitization process verified? Is there a manual review step?
3. Are there plans to support other platforms (e.g., macOS, web)?
4. What is the intended pricing model or monetization strategy?
5. Has the tool been tested with real users in engineering environments?
6. How does Sanitizer handle edge cases like scanned documents or complex metadata?
7. Is there a roadmap for expanding document formats or AI integration?