What problem does it solve? Bibliometric datasets exported from platforms like Web of Science, CNKI, Scopus, and PubMed use completely different field structures, encodings, and record separators. Without identifying the source first, every downstream parsing and cleaning step fails or produces corrupted results. ## Core Features & Use Cases - Platform Fingerprint Detection: Identifies Web of Science, CNKI, Scopus, PubMed/MEDLINE, Dimensions, IEEE Xplore, ProQuest, Wanfang, VIP, and Google Scholar by matching file signatures and field labels. - Encoding and Format Analysis: Detects file encoding (UTF-8, GBK, UTF-16, etc.), line endings, record separators, and export format variants (plain text, CSV, Excel, BibTeX, RIS, EndNote, RefWorks). - Structural Validation: Samples records from the head, middle, and tail of the file to verify field completeness, separator consistency, and detect mixed-source merged datasets. - Use Case: You receive a CNKI export file with GBK encoding and Chinese field labels. The skill identifies it as a CNKI custom export, reports the field mapping, and routes the correct parsing rules to downstream cleaning skills. ## Quick Start Analyze the attached literature data file and tell me which platform it was exported from, its format, and its encoding.
