Converts audio recordings and PDF documents into structured markdown for qualitative research and analysis.
The Interview Ingest skill facilitates the critical first step of qualitative research by streamlining the transformation of raw data into AI-ready formats. It provides a dual-tier approach for processing complex PDF documents and audio recordings, utilizing advanced VLM-powered parsing through MinerU for high accuracy or manual workflows for flexible data import. By organizing outputs into standardized markdown with speaker labels and automated data inventory tracking, it creates a robust foundation for deep qualitative coding and thematic analysis within the Claude Code environment.
Key Features
01Automated markdown generation with speaker labeling and timestamping support
022 GitHub stars
03Centralized data-inventory.json tracking for project-wide source management
04Multi-tier PDF parsing with MinerU VLM integration for 90%+ accuracy on tables and figures
05Standardized audio transcription workflows for Otter.ai, Rev.com, and Whisper integration
06Batch processing scripts for high-volume interview and document ingestion
Use Cases
01Converting academic research papers with complex tables into analyzable markdown
02Preparing large sets of user interview recordings for thematic qualitative coding
03Digitizing legacy PDF archives into a structured repository for AI-assisted literature reviews
