What problem does it solve? Before cleaning a bibliographic dataset exported from platforms like Web of Science or Scopus, researchers need to know what the data actually contains: how many records exist, which fields are missing, what document types and languages are present, and whether duplicates or retracted papers lurk inside. This Skill performs a full structural scan and produces a data profile report answering those questions. ## Core Features & Use Cases - Field Completeness Audit: Counts populated versus missing values for every field and classifies fields as core, key, auxiliary, index, or redundant. - Distribution Analysis: Reports document type, language, and publication year distributions, flagging non-English homograph noise and anomalous years. - Quality & Duplicate Detection: Identifies exact duplicates by UT/DOI/PMID, near-duplicates by title similarity, empty or truncated records, and retraction or correction markers. - Use Case: After S01 detects that a file is a Web of Science plain-text export, run this Skill to generate a data analysis report showing 95% abstract completeness, 12 duplicate records, and 3 retracted articles before proceeding to cleaning. ## Quick Start Analyze the bibliographic data file at the given path and generate a full data profile report covering field completeness, document types, languages, years, and duplicates.
