Extract, clean, transform, analyze, and export CSV data using natural language commands on large files with auto-detection of delimiter and formats.
# csv-data-extractor Transform, clean, and analyze CSV data using natural language commands. Upload a CSV file and describe what you want to do with it. ## Usage ``` Analyze this CSV and summarize [file] Extract [specific data] from [file] Filter rows where [condition] and export [file] Join [file1] and [file2] on [column] Convert this data to [format: JSON/Excel/HTML table] ``` ## What It Does - **Data profiling** — row count, column types, null values, unique counts, value distributions - **Filtering** — select rows matching conditions (e.g., "where status = 'active' and amount > 1000") - **Column operations** — rename, reorder, drop, split, merge, type cast - **Aggregations** — GROUP BY with SUM, AVG, COUNT, MIN, MAX, pivot tables - **Joins** — LEFT/RIGHT/INNER join on specified columns - **Transformations** — date parsing, string operations, currency formatting, unit conversions - **Export** — output to CSV, JSON, Excel (.xlsx), HTML table, or Markdown ## Input A CSV file path or paste CSV content directly in the conversation. ## Examples - "Show me the top 10 customers by revenue from this sales data" - "Filter out all rows where email is blank, then export only name/email/phone" - "Add a column called 'full_name' that combines first_name and last_name" - "Group by department and show average salary per department" - "Convert this data to JSON with nested objects by category" ## Notes - Handles large files (100k+ rows) by processing in chunks - Auto-detects delimiter (comma, tab, semicolon, pipe) - Detects date formats automatically - Preserves original file; always exports to new file - Encodings supported: UTF-8, Latin-1, Windows-1252
don't have the plugin yet? install it then click "run inline in claude" again.