Agent Skills: Bilig WorkPaper Agent Skill

Use Bilig WorkPaper for spreadsheet-style formulas in Node.js services and AI agent tools without driving Excel or browser spreadsheet UIs.

UncategorizedID: proompteng/bilig/bilig-workpaper

Install this agent skill to your local

pnpm dlx add-skill https://github.com/proompteng/bilig/tree/HEAD/skills/bilig-workpaper

Skill Files

Browse the full folder contents for bilig-workpaper.

Download Skill

Loading file tree…

skills/bilig-workpaper/SKILL.md

Skill Metadata

Name
bilig-workpaper
Description
Use Bilig WorkPaper for spreadsheet-style formulas in Node.js services and AI agent tools without driving Excel or browser spreadsheet UIs.

Bilig WorkPaper Agent Skill

Use this skill when an agent needs spreadsheet-style formulas through files, terminal commands, TypeScript, HTTP routes, or MCP tools instead of Excel UI automation.

When To Use

Use Bilig for:

  • backend pricing, quote, payout, budget, import-validation, or forecast logic;
  • formula readback after writing input cells;
  • deterministic cell-address workflows for coding agents;
  • MCP clients that can run a stdio server;
  • reduced XLSX formula/import bug reports.

Do not use it for manual spreadsheet editing, Office macros, VBA, pivots, charts, COM automation, or exact Excel desktop behavior unless the task is an explicit compatibility comparison.

Command Safety

Do not build shell commands by concatenating user text. Treat the commands below as literal templates, validate workbook paths before use, and reject values containing newlines, backticks, $(, ;, &, |, <, or >. Prefer MCP client command plus args arrays or direct TypeScript calls when inserting user-provided paths or cell references.

MCP Path

When the agent host supports stdio MCP, start the file-backed WorkPaper server with an argument array, not a shell-concatenated string:

{
  "command": "npm",
  "args": [
    "exec",
    "--package",
    "@bilig/headless@0.23.3",
    "--",
    "bilig-workpaper-mcp",
    "--workpaper",
    "./pricing.workpaper.json",
    "--init-demo-workpaper",
    "--writable"
  ]
}

Useful MCP tools:

  • list_sheets
  • read_range
  • read_cell
  • set_cell_contents
  • get_cell_display_value
  • export_workpaper_document
  • validate_formula

After each write, read the dependent output cell and export the WorkPaper document as persistence proof.

TypeScript Path

Use @bilig/headless directly when workbook logic belongs in a service, worker, test, or route:

import { WorkPaper, exportWorkPaperDocument, serializeWorkPaperDocument } from "@bilig/headless";

const workbook = WorkPaper.buildFromSheets({
  Inputs: [
    ["Metric", "Value"],
    ["Customers", 20],
    ["Average revenue", 1200],
  ],
  Summary: [
    ["Metric", "Value"],
    ["Revenue", "=Inputs!B2*Inputs!B3"],
  ],
});

const inputs = workbook.getSheetId("Inputs");
const summary = workbook.getSheetId("Summary");
if (inputs === undefined || summary === undefined) {
  throw new Error("Workbook is missing required sheets");
}

workbook.setCellContents({ sheet: inputs, row: 1, col: 1 }, 32);
const revenue = workbook.getCellDisplayValue({ sheet: summary, row: 1, col: 1 });
const saved = serializeWorkPaperDocument(exportWorkPaperDocument(workbook, { includeConfig: true }));

console.log({ revenue, savedBytes: saved.length });

Required Proof

A successful response should include:

  • exact edited sheet and A1 cell;
  • before values for relevant inputs and dependent outputs;
  • after values read from the recalculated workbook;
  • persistence evidence from serialized or exported WorkPaper state;
  • limitations for unsupported formulas or Excel-only features.

If any proof step fails, report the blocker instead of claiming the workbook was updated.

References

  • Docs map: https://proompteng.github.io/bilig/llms.txt
  • Agent handbook: https://proompteng.github.io/bilig/headless-workpaper-agent-handbook.html
  • MCP guide: https://proompteng.github.io/bilig/mcp-workpaper-tool-server.html
  • Formula clinic: https://proompteng.github.io/bilig/formula-bug-clinic.html
  • Repository: https://github.com/proompteng/bilig