Document Feature List Extractor

Problem Addressed

Product managers often need to turn bid documents, technical agreements, solution documents, and meeting notes into reviewable feature lists. Source material may hide hierarchy inside paragraphs, tables, and numbered lists, which can lead to missing items, inconsistent levels, or mismatched descriptions.

This skill converts document content into a structured feature list. It reads .docx, .pdf, .txt, .md, or pasted text, identifies Level 1, Level 2, Level 3 features, and feature descriptions, and generates an Excel workbook with merged cells, usually named feature_list_YYYYMMDD.xlsx.

How It Works and Limitations

The main workflow is:

• Read the document: supports Word, PDF, plain text, Markdown, and pasted content.
• Analyze the feature structure: separate modules, submodules, and concrete feature points into levels.
• Extract feature details: capture feature names, hierarchy relationships, and descriptions as complete as possible.
• Generate Excel: use openpyxl to create headers, fill data, set column widths, and merge repeated values under the same parent.

The documented workflow supports up to three feature levels; deeper levels require extension. If the source document is poorly structured, the extracted list may need manual cleanup. This skill is best used to produce a first draft of a feature list, not to replace manual confirmation of boundaries, acceptance criteria, or pricing scope.