Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Spreadsheet creation, format conversion (ODS/XLSX/CSV), formulas, data automation with LibreOffice Calc.
.claude/skills/lingxling-calc/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 54% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 97% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 75% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 53% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 25% | 0% |
LibreOffice Calc skill for creating, editing, converting, and automating spreadsheet workflows using the native ODS (OpenDocument Spreadsheet) format.
Use this skill when:
bashsoffice --calc template.ods
pythonimport uno def create_spreadsheet(): local_ctx = uno.getComponentContext() resolver = local_ctx.ServiceManager.createInstanceWithContext( "com.sun.star.bridge.UnoUrlResolver", local_ctx ) ctx = resolver.resolve( "uno:socket,host=localhost,port=8100;urp;StarOffice.ComponentContext" ) smgr = ctx.ServiceManager doc = smgr.createInstanceWithContext("com.sun.star.sheet.SpreadsheetDocument", ctx) sheets = doc.getSheets() sheet = sheets.getByIndex(0) cell = sheet.getCellByPosition(0, 0) cell.setString("Hello from LibreOffice Calc!") doc.storeToURL("file:///path/to/spreadsheet.ods", ()) doc.close(True)
pythonimport ezodf doc = ezodf.newdoc('ods', 'spreadsheet.ods') sheet = doc.sheets[0] sheet['A1'].set_value('Hello') sheet['B1'].set_value('World') doc.save()
bash# ODS to XLSX soffice --headless --convert-to xlsx spreadsheet.ods # ODS to CSV soffice --headless --convert-to csv spreadsheet.ods # ODS to PDF soffice --headless --convert-to pdf spreadsheet.ods # XLSX to ODS soffice --headless --convert-to ods spreadsheet.xlsx # Batch convert for file in *.ods; do soffice --headless --convert-to xlsx "$file" done
pythonimport uno def create_formula_spreadsheet(): local_ctx = uno.getComponentContext() resolver = local_ctx.ServiceManager.createInstanceWithContext( "com.sun.star.bridge.UnoUrlResolver", local_ctx ) ctx = resolver.resolve( "uno:socket,host=localhost,port=8100;urp;StarOffice.ComponentContext" ) smgr = ctx.ServiceManager doc = smgr.createInstanceWithContext("com.sun.star.sheet.SpreadsheetDocument", ctx) sheet = doc.getSheets().getByIndex(0) sheet.getCellByPosition(0, 0).setDoubleValue(100) sheet.getCellByPosition(0, 1).setDoubleValue(200) cell = sheet.getCellByPosition(0, 2) cell.setFormula("SUM(A1:A2)") doc.storeToURL("file:///path/to/formulas.ods", ()) doc.close(True)
bashsoffice --headless soffice --headless --convert-to <format> <file> soffice --calc # Calc
bashpip install ezodf # ODS handling pip install odfpy # ODF manipulation pip install pandas # Data analysis
bashkillall soffice.bin soffice --headless --accept="socket,host=localhost,port=8100;urp;"
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 14,839 | 19,238 | +30% | 1 | 1 | 0% | 2,319 | 3,605 | +55% | 0 | 0 | — |
case-02 | pass→pass | 41,163 | 12,735 | -69% | 1 | 1 | 0% | 2,118 | 3,256 | +54% | 0 | 0 | — |
case-03 | fail→pass | 11,965 | 12,450 | +4% | 1 | 1 | 0% | 1,896 | 2,917 | +54% | 0 | 0 | — |
case-04 | fail→pass | 6,110 | 4,027 | -34% | 1 | 1 | 0% | 1,053 | 2,078 | +97% | 0 | 0 | — |
case-05 | fail→pass | 5,778 | 3,164 | -45% | 1 | 1 | 0% | 1,078 | 1,888 | +75% | 0 | 0 | — |
case-06 | fail→pass | 7,397 | 3,052 | -59% | 1 | 1 | 0% | 1,124 | 1,725 | +53% | 0 | 0 | — |
case-07 | fail→pass | 12,412 | 10,704 | -14% | 1 | 1 | 0% | 2,189 | 2,743 | +25% | 0 | 0 | — |
case-08 | fail→pass | 18,779 | 4,181 | -78% | 1 | 1 | 0% | 2,047 | 1,898 | -7% | 0 | 0 | — |
case-09 | pass→pass | 7,514 | 3,493 | -54% | 1 | 1 | 0% | 1,369 | 2,094 | +53% | 0 | 0 | — |
case-10 | fail→fail | 8,322 | 13,023 | +56% | 1 | 1 | 0% | 1,228 | 3,115 | +154% | 0 | 0 | — |
case-11 | pass→pass | 9,543 | 7,032 | -26% | 1 | 1 | 0% | 1,316 | 2,596 | +97% | 0 | 0 | — |
case-12 | fail→pass | 8,900 | 9,456 | +6% | 1 | 1 | 0% | 1,477 | 2,609 | +77% | 0 | 0 | — |
case-13 | fail→pass | 10,759 | 4,615 | -57% | 1 | 1 | 0% | 1,509 | 1,982 | +31% | 0 | 0 | — |
case-14 | fail→pass | 17,932 | 5,253 | -71% | 1 | 1 | 0% | 2,644 | 1,923 | -27% | 0 | 0 | — |
case-15 | fail→pass | 31,846 | 12,482 | -61% | 1 | 1 | 0% | 1,722 | 3,195 | +86% | 0 | 0 | — |
case-16 | pass→pass | 11,955 | 7,070 | -41% | 1 | 1 | 0% | 2,139 | 2,398 | +12% | 0 | 0 | — |
case-17 | pass→pass | 11,545 | 3,235 | -72% | 1 | 1 | 0% | 1,454 | 1,793 | +23% | 0 | 0 | — |
case-18 | pass→pass | 12,509 | 5,451 | -56% | 1 | 1 | 0% | 1,782 | 2,005 | +13% | 0 | 0 | — |
case-19 | pass→pass | 17,721 | 10,176 | -43% | 1 | 1 | 0% | 2,398 | 2,649 | +10% | 0 | 0 | — |
case-20 | fail→fail | 25,106 | 21,495 | -14% | 1 | 1 | 0% | 3,771 | 4,843 | +28% | 0 | 0 | — |
case-21 | fail→fail | 19,764 | 23,360 | +18% | 1 | 1 | 0% | 3,622 | 5,739 | +58% | 0 | 0 | — |
case-22 | fail→fail | 17,851 | 15,871 | -11% | 1 | 1 | 0% | 2,749 | 3,865 | +41% | 0 | 0 | — |
case-23 | pass→pass | 8,290 | 5,858 | -29% | 1 | 1 | 0% | 1,134 | 2,168 | +91% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +43 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.