Research Quality Audit
Audits research repositories for content depth, identifies placeholder stubs, and automatically expands shallow documentation using AI agents.
The Research Quality Audit skill provides a specialized cognitive workflow for maintaining high-fidelity research corpuses within Claude Code. It scans your documentation for 'stubs'—analysis files written only from abstracts rather than full source material—and categorizes entries into quality tiers based on metrics like line count, section density, and quote frequency. When used with its auto-fix capability, it dispatches expansion agents to read source PDFs and rewrite shallow documents into comprehensive analysis reports, ensuring your knowledge base remains substantive, verified, and reliable.
Key Features
01Multi-tier quality classification ranging from 'Full' analysis to 'Skeleton' placeholders
02Source verification to ensure referenced PDFs and extracted text files are present
03146 GitHub stars
04Automated detection of docs written from abstracts rather than full-text sources
05Auto-fix functionality that dispatches agents to autonomously expand shallow content
06Comprehensive statistical reporting including line count distribution and depth metrics
Use Cases
01Validating the quality of a research corpus after a large-scale batch induction
02Identifying and fixing 'shallow stubs' in documentation that lacks methodology or results
03Automating the expansion of placeholder research notes into full-length analysis reports