{"id":67362,"plugin_id":"plugin_asdk_app_6ac12ac1fde881918ec1761f93098132","kind":"skill","collection_source":"plugin_package","comparison_source":null,"observed_at":"2026-10-28T13:07:49.474Z","digest":"c0a5df328062b3ea49d594554a7814162120e5b726109e6c91985fbe17578516","against":null,"payload":{"description":"Spreadsheet merge tools. join_spreadsheets joins two CSV/Excel files on an ID column (VLOOKUP job) with counted key clean-ups (spaces, case, leading zeros, Excel number formats), unmatched-row reasons and duplicate checks; combine_spreadsheets stacks up to 24 files with drifting column names into one sheet and lists every column decision.","included_files":[{"relative_path":"agents/openai.yaml","size_in_bytes":536}],"name":"sheet-join","skill_md_contents":"---\nname: sheet-join\ndescription: Spreadsheet merge tools. join_spreadsheets joins two CSV/Excel files on an ID column (VLOOKUP job) with counted key clean-ups (spaces, case, leading zeros, Excel number formats), unmatched-row reasons and duplicate checks; combine_spreadsheets stacks up to 24 files with drifting column names into one sheet and lists every column decision.\n---\n\n# Zovo Spreadsheet Merge\n\nUse `join_spreadsheets` when the user wants to VLOOKUP, match or merge two files side by side on an ID: \"vlookup two spreadsheets\", \"merge orders with customers by customer ID\", \"why don't these IDs match\". Pass the main file (rows to keep) as `left_file` and the lookup file as `right_file`. One workbook with both tables on two sheets can go alone in `left_file`. Without files, paste each table as CSV text in `left_table` / `right_table`, header row first.\n\n1. Leave `left_key`/`right_key` out unless the user names the key; the tool explains the key it picked. Use a list for a two-column key.\n2. `join_type` left (default) keeps every left row; inner = only matches; full = everything.\n3. If the user says IDs are case-sensitive, or that 0123 and 123 are different codes, pass `skip_normalizations` ([\"case\"], [\"zeros\"]).\n4. Reply with the match rate first, then each clean-up with its row count (these are the rows a plain VLOOKUP misses), the unmatched reasons, duplicate-key warnings, and the two download links (they expire in about an hour). Quote keys exactly as the tool shows them.\n\nUse `combine_spreadsheets` to stack or append files: \"combine 12 monthly CSVs\", \"combine excel files with different column names\", \"combine all sheets into one\". Pass all files in `files` (up to 24), or pasted tables in `tables`. Read back the column decisions, list every fuzzy match under \"please confirm\", and offer `column_map` (e.g. {\"Amt\":\"Amount\"}) to change any. Pass `date_order` only when the user states the date format.\n\nDo not use these tools for mail merge (letters, labels, emails), for comparing two drafts of a document, or for explaining Excel formulas.\n"},"changes":[],"summary":"First saved snapshot. No earlier version is available for comparison.","summary_kind":"deterministic","summary_metadata":{}}