What is the benefit of the data-inventory.json file?

It creates a programmatic audit trail that tracks original files, conversion tools used, and metadata, ensuring transparency in your research data pipeline.

What file formats does Qual-Ingest support?

Qual-Ingest supports a wide range of formats including audio (MP3/WAV), PDFs, DOCX, XLSX, PPTX, images, and web content.

Do I need an API key to use this skill?

No, an API key is only required for Tier 1 features like MinerU's high-accuracy PDF parsing. Tier 2 supports manual conversion and common OCR tools without an API.

How does it handle interview transcripts?

It provides a standardized workflow for importing transcripts from services like Otter.ai or Rev.com, adding speaker labels, and organizing them for analysis.

Can I use this for complex academic papers?

Yes, when used with a MinerU API key, the skill uses VLM-powered parsing to extract tables, formulas, and figures with over 90% accuracy.

Qualitative Data Ingestion

Name: Qualitative Data Ingestion
Author: linxule

bylinxule

•

データサイエンスとML

Converts audio recordings, PDFs, and diverse document formats into structured markdown for qualitative research and AI-assisted analysis.

Qual-Ingest is a specialized capability designed for researchers and analysts to streamline the preparation of raw qualitative data within the Claude Code environment. It facilitates the conversion of interview recordings and complex PDFs into clean, analyzable markdown files, offering a two-tier system that supports high-accuracy VLM parsing via MinerU or manual workflows. By automating the transition from raw media to structured data and maintaining a detailed data inventory, it establishes a rigorous foundation for multi-stage qualitative coding and thematic synthesis.

主な機能

01Support for multi-format conversion including DOCX, XLSX, images, and web content

02Standardized output organization optimized for subsequent qualitative coding stages

03Advanced PDF parsing using MinerU VLM-powered extraction for tables and figures

042 GitHub stars

05Automated data inventory management with JSON tracking for audit trails

06Structured audio-to-markdown workflows for interview transcripts and speaker labeling

ユースケース

01Transcribing and formatting academic interview recordings for thematic analysis

02Converting complex research papers and PDFs with dense tables into clean markdown

03Organizing diverse raw data sources into a centralized repository for AI-assisted research

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add linxule/interpretive-orchestration qual-ingest

For use in Claude.ai and ChatGPT

Download Skill