portfolio-ingest

To extract portfolio entities from product management documents, you upload files like markdown, docx, pptx, xlsx, or pdf to the uploads directory. The Skill parses these documents to identify and classify products, features, and markets.

Automating product portfolio data ingestion from existing files is best handled by parsing uploaded documents against an existing portfolio.json file. This process extracts new entities, prevents duplicates, and moves processed files to a dedicated directory.

Yes, you can extract product features and markets from PDF and Word documents. The Skill supports ingestion from markdown, docx, pptx, xlsx, and pdf formats, leveraging document-skills to process non-markdown files for entity extraction.

Yes, an existing portfolio.json file is analyzed during ingestion to prevent duplicate entity creation. The Skill cross-references extracted products, features, and markets against this file before creating new entity JSON files.

Document processing handles extracted portfolio entities by creating new JSON files for them. After successful extraction and classification, the original uploaded documents are automatically moved to the uploads/processed/ directory.