literature-export

Export existing PDF, Zotero, or WoS records to Excel and CSV while preserving sources, evidence scope, and missing values.

SKILL.md

GitHub

Literature Table Export

Read shared contract. Use the spreadsheet skill when generating workbooks.

Use prepare_export.py --input STRUCTURED_JSON_OR_WOS_DIRECTORY --output OUTPUT to normalize records into tables.json and CSV files. Supported inputs are library.json, screening-records.json and a WoS analysis directory. Preserve evidence scope and null values; do not write summaries for records only screened by title.

Run node export_tables.mjs OUTPUT NODE_RUNTIME_DIRECTORY for XLSX when @oai/artifact-tool is available. Inspect every sheet preview and verify row counts. Otherwise deliver CSV and explain XLSX was not produced. Keep all document records; a selected-reference sheet must state that it contains only the network selection, with full references retained in the WoS CSV/JSON.

If no structured input exists, route first to the appropriate PDF, Zotero or WoS module. Exporting all records does not mean all documents have been read in full.