xlsx-for-ai 2.6.0 → 2.8.0

package/mcp.js

@@ -564,6 +564,44 @@ const TOOLS = [
     },
   },
 
+  {
+    name: 'xlsx_charts',
+    description:
+      'xlsx-for-ai — read, write, diff, redact, supervise .xlsx files locally.\n' +
+      'This tool: list every chart in a workbook with type (bar / line / pie / scatter / area / doughnut / radar / stock / surface / bubble), title, axis titles, and per-series formula refs (the cell ranges the chart pulls from). Sheet attribution via the OOXML drawing rel chain.\n' +
+      'No other tool can do this: ExcelJS doesn\'t expose charts at all (read or write); pandas drops them entirely; openpyxl reads charts but in a deeply-nested object form unsuitable for LLM consumption. xlsx_charts gives you the chart contract — "Sheet2 has a bar chart titled Q4 Revenue plotting Sheet1!B2:B10 against Sheet1!A2:A10" — without rendering anything.\n\n' +
+      'USE WHEN: documenting a financial model / dashboard for an LLM that needs to know "what does this workbook visualize, and from which cells?". Or auditing a workbook for chart-data drift after a refactor (chart still points at old range?). ' +
+      'Free tier — counts against the 10k/mo cap.\n\n' +
+      'DO NOT USE WHEN: you want to RENDER the chart as an image (this tool returns the chart spec, not pixels). Or you only need cell values (use xlsx_read).',
+    inputSchema: {
+      type: 'object',
+      properties: {
+        file_path: { type: 'string', description: 'Absolute path to the .xlsx file.' },
+        sheet: { type: 'string', description: 'Optional: restrict to a specific sheet.' },
+      },
+      required: ['file_path'],
+    },
+  },
+
+  {
+    name: 'xlsx_protection',
+    description:
+      'xlsx-for-ai — read, write, diff, redact, supervise .xlsx files locally.\n' +
+      'This tool: surface every protection setting in a workbook so an agent knows what it can and cannot edit. Workbook-level (lockStructure, lockWindows), per-sheet (protected? password? hidden state?), per-action allow/block list (formatCells, sort, insertRows, pivotTables, etc.), and per-cell unlocked / hidden samples — these are the cells a human would actually be allowed to type into when the sheet is otherwise read-only.\n' +
+      'No other tool can do this: pandas drops protection metadata entirely; openpyxl exposes the bool but no normalization. xlsx_protection reads sheetProtection action attrs directly from the OOXML zip (workaround for ExcelJS stripping them on round-trip).\n\n' +
+      'USE WHEN: an agent is about to suggest edits to a workbook and you want to fail fast on cells / sheets the user can\'t change anyway. Or auditing a "submitted form" workbook to see which inputs the form-author intended to be fillable. ' +
+      'Free tier — counts against the 10k/mo cap.\n\n' +
+      'DO NOT USE WHEN: just reading values (use xlsx_read). Or trying to BREAK protection (this tool surfaces what\'s locked; it does not unlock).',
+    inputSchema: {
+      type: 'object',
+      properties: {
+        file_path: { type: 'string', description: 'Absolute path to the .xlsx file.' },
+        sheet: { type: 'string', description: 'Optional: restrict to a specific sheet.' },
+      },
+      required: ['file_path'],
+    },
+  },
+
   {
     name: 'xlsx_styles',
     description:

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "xlsx-for-ai",
   "mcpName": "io.github.senoff/xlsx-for-ai",
-  "version": "2.6.0",
+  "version": "2.8.0",
   "description": "The MCP server that makes LLMs reliable on real-world Excel spreadsheets. Thin npm client over a hosted API — read, write, diff, redact, and supervise .xlsx files from any MCP-aware agent.",
   "main": "index.js",
   "bin": {