What problem does it solve? Working with Lark (Feishu) spreadsheets programmatically is error-prone: agents guess wrong command names, write numbers as text, skip read-back verification, and mistranslate Excel formulas. This Skill provides a complete, verified command vocabulary and editing discipline for reliably creating and modifying online spreadsheets through the lark-cli tool. ## Core Features & Use Cases - Full spreadsheet lifecycle: Create workbooks, manage sheets (insert/delete/rename/copy), read and write cells with values, formulas, styles, comments, and embedded images, plus find-and-replace and batch updates. - Data objects and visualization: Build native charts, pivot tables, conditional formats, filters, filter views, sparklines, and floating images as real spreadsheet objects that stay linked to source data. - Formula correctness pipeline: Translate Excel formulas into Lark-executable semantics (ARRAYFORMULA, MAP/LAMBDA, native array functions) and verify zero-error results with the formula-verify self-check. - Use Case: A user asks to build a financial model in a Lark spreadsheet: create the workbook with typed data, write DCF formulas, add a pivot table for segment summary, generate a trend chart, apply conditional formatting, and confirm every formula computes without errors. ## Quick Start Ask the AI to create a new Lark spreadsheet with a sales table, add a pivot table summarizing revenue by region, and verify all formulas compute correctly.