plugin-document-parser
v1.1.0
Published
Intercept and route AI chat file attachments through configurable internal or external OCR/parse providers.
Readme
plugin-document-parser
Overview
Intercept and route AI chat file attachments through configurable internal or external OCR/parse providers.
Features
- Pluggable Parsers: Route documents to MarkItDown, Unstructured.io, DocPixie, or custom OCR engines.
- Format Normalization: Standardizes diverse file types into clean Markdown for LLM consumption.
- Pipeline Management: Handle caching, error recovery, and parsing statuses.
Usage
- Enable the plugin.
- Go to Document Parser Settings.
- Configure your preferred parsing engine for specific file types (e.g., use MarkItDown for Office files, DocPixie for PDFs).
- Whenever a user uploads a file in the AI chat, it will automatically pass through this pipeline.
