Add PowerPoint (.pptx) slide text extraction to RAG module
Add PowerPoint (.pptx) slide text extraction to RAG module: a task in LegoFlow-SWE (Harbor dataset). Problem The RAG module currently provides readers for PDF, Word, Excel, and plain text, but it cannot ingest PowerPoint presentations. Users who want to use slide content in retrieval-augmented…
The task
**Problem** The RAG module currently provides readers for PDF, Word, Excel, and plain text, but it cannot ingest PowerPoint presentations. Users who want to use slide content in retrieval-augmented generation workflows must manually convert .pptx files first.
Part of Lego-X/LegoFlow-SWE.