What problem does it solve? Working with Feishu (Lark) spreadsheets programmatically is error-prone: agents guess wrong command names, lose number/date types when writing data, break formulas during Excel migration, and leave half-finished edits. This Skill provides a complete, verified command map and strict workflow rules for operating Lark spreadsheets through lark-cli. ## Core Features & Use Cases - Full spreadsheet lifecycle: create workbooks, import/export xlsx/csv, manage sheets, rows, columns, merges, freeze panes, and read/write cell values, formulas, styles, comments, and embedded images. - Data objects and visualization: create and maintain charts, pivot tables, conditional formats, filters, filter views, sparklines, and floating images as real spreadsheet objects that update with source data. - Formula and style governance: authoritative rules for translating Excel formulas to Feishu (ARRAYFORMULA, MAP/LAMBDA, date differences, unsupported functions) plus visual styling standards and atomic batch updates with read-back verification. - Use Case: A user asks to build a sales summary: the Skill reads the workbook structure, writes typed data with +table-put, creates a pivot table for group aggregation, adds a chart excluding total rows, and verifies results by reading back the affected ranges. ## Quick Start Ask the AI to read the Lark spreadsheet at your /sheets/ URL and summarize or edit its data using the lark-sheets skill.
