Montscan is an automated document processing system designed to streamline the ingestion and organization of scanned documents. It targets users who receive scans via FTP and need intelligent file naming and cloud storage integration, offering automated processing to save manual effort. The system leverages AI for both document analysis and filename generation, integrating seamlessly with WebDAV-compatible cloud storage solutions.

How It Works

Montscan operates by receiving scanned documents through an integrated FTP server. Upon receipt, it utilizes Ollama vision models to analyze the document's content. Subsequently, it employs Ollama AI to generate descriptive filenames, with the capability to support different languages. Finally, processed documents are automatically uploaded to cloud storage via WebDAV, supporting services like Nextcloud and ownCloud.

Quick Start & Requirements

• Primary install/run command: go build -o montscan . followed by ./montscan. Docker Compose is also supported (docker-compose up -d).
• Prerequisites: Go 1.24+, Poppler (pdftoppm) or ImageMagick for PDF conversion, Ollama with a vision model (e.g., llava, llama3.2-vision - see https://ollama.ai/), and an optional WebDAV server.
• Estimated setup time: Requires installation of Go, PDF/image tools, Ollama, and potentially configuring a WebDAV server. Docker simplifies deployment.
• Relevant pages: GitHub repository (implied), Ollama installation guide.

Highlighted Details

• Achieves a 97.5% success rate on a small test set of 1000 documents.
• Features an FTP server for receiving scans from network scanners.
• Integrates Vision AI processing via Ollama for document analysis.
• Provides AI-powered naming for generating descriptive filenames (defaulting to English).
• Supports WebDAV integration for automatic uploads to cloud storage.
• Offers Docker support for simplified deployment via Docker Compose.

Maintenance & Community

No specific details regarding notable contributors, sponsorships, community channels (like Discord/Slack), or a public roadmap are provided in the README. Support is available by opening an issue on GitHub.

Licensing & Compatibility

Licensed under the MIT License. This license is generally permissive and compatible with commercial use and closed-source linking.

Limitations & Caveats

The project is explicitly stated as "not fully production-ready and is currently in active development." The reported 97.5% success rate was achieved on a "small test set." While the configuration supports a LANGUAGE variable, the README description mentions French filenames while the default configuration is English.