This project provides an automated document analysis and tagging system for Paperless-ngx, leveraging various AI models and APIs. It targets users of Paperless-ngx seeking to enhance document organization and retrieval through AI-powered metadata assignment, offering both automated and manual analysis modes, along with an interactive chat feature for document querying.

How It Works

Paperless-AI integrates with Paperless-ngx to automatically scan for new documents. It then utilizes AI models, accessed via APIs like OpenAI, Ollama, and others, to analyze document content. The system can automatically assign titles, tags, document types, and correspondents based on the AI's analysis, with options for custom rules and selective tag application.

Quick Start & Requirements

• Installation: Full Docker support is provided. For local development, install dependencies with npm install and start the server with npm run test.
• Prerequisites: Requires Paperless-ngx, Node.js, and access to AI model APIs (OpenAI, Ollama, etc.).
• Links: Installation instructions are available on the project's Wiki.

Highlighted Details

• Supports multiple AI services including OpenAI, Ollama (Mistral, Llama, Phi 3, Gemma 2), DeepSeek.ai, OpenRouter.ai, and more.
• Features an interactive chat interface for querying document content with AI.
• Offers both automated document processing and a manual analysis mode via a /manual endpoint.
• Provides customization options for processing rules and tag assignment.

Maintenance & Community

• The project is open for contributions via pull requests.
• Users can report issues or ask questions in the Issues section.

Licensing & Compatibility

• Licensed under the MIT License.
• Compatible with commercial use and closed-source linking due to its permissive MIT license.

Limitations & Caveats

The project is primarily designed for integration with Paperless-ngx, and its functionality is dependent on the availability and configuration of external AI services.