HyperFormula Alternative For Node WorkPaper Automation

May 29, 2026 ยท View on GitHub

Status: public comparison guide for @bilig/headless.

This page is for developers already evaluating HyperFormula or another headless spreadsheet engine. It is intentionally not a takedown. HyperFormula is the established TypeScript engine in this category; its official README positions it as a headless spreadsheet for business web apps with formula evaluation, CRUD operations, undo/redo, clipboard support, sorting, Node.js support, and a GPLv3 or commercial license.

@bilig/headless is worth evaluating when the workload is closer to a service-side WorkPaper runtime: formula-backed business logic, structural edits, agent writeback, persistence, restore, and auditable benchmark evidence from the same repository.

Short Version

Use HyperFormula when you want a mature, UI-independent formula calculation engine with broad spreadsheet-function coverage and commercial support from the Handsontable team.

Use @bilig/headless when you need a MIT-licensed Node package that treats the workbook as a service object: build it, mutate it, read formulas and values, persist it, restore it, and verify agent-style edits without opening a browser grid.

For the broader engine choice, start with the headless spreadsheet engine use-case chooser.

Comparison Surface

QuestionHyperFormula@bilig/headless
Primary shapeHeadless spreadsheet formula engineHeadless WorkPaper workbook facade
Runtime targetBrowser or Node.jsNode services, tests, agents, and local runtime automation
License postureGPLv3 or commercial licenseMIT
API orientationSpreadsheet-engine instance with formula evaluation and workbook operationsWorkPaper object with formula evaluation, structural edits, persistence helpers, history, and readback
Agent workflow fitPossible, but the project is not specifically packaged around agent writeback proofsFirst-class evaluation path includes an agent writeback demo with persistence and restored readback
Benchmark claim in this repoExternal comparison targetChecked-in WorkPaper-vs-HyperFormula artifact records 100/100 mean wins on scorecard-eligible comparable workloads
CaveatStrong default engine, with its own licensing and integration modelFormula inventory is complete for the tracked Office surface, but arbitrary Excel edge cases remain evidence-scoped

What The Benchmark Says

The current checked-in artifact is:

packages/benchmarks/baselines/workpaper-vs-hyperformula.json

The short benchmark explainer is:

docs/what-workpaper-benchmark-proves.md

The current public claim is narrow:

  • 100/100 mean wins on scorecard-eligible comparable workloads
  • 73/73 public-lane mean wins
  • 27/27 holdout-lane mean wins
  • 100/100 mean+p95 wins on scorecard-eligible comparable workloads
  • an overall p95 geomean lead with the named worst p95 row still visible

The verification command is:

pnpm workpaper:bench:competitive:check

This does not prove that bilig is faster at every possible spreadsheet task. It does not prove full Excel compatibility. It proves the checked-in WorkPaper runtime claim for the current comparable headless workload scorecard.

Try The Package

Use the published package when you want a quick local evaluation:

mkdir bilig-headless-eval
cd bilig-headless-eval
npm init -y
npm pkg set type=module
npm install @bilig/headless

Use the maintained example when you want an end-to-end proof:

git clone https://github.com/proompteng/bilig.git
cd bilig
pnpm --dir examples/headless-workpaper install --ignore-workspace
pnpm --dir examples/headless-workpaper run start
pnpm --dir examples/headless-workpaper run agent:verify

The agent verifier records the assumption cells changed, checks dependent formula readback, persists the workbook, restores it, and verifies the restored values.

When To Choose bilig First

  • You need a workbook object in a Node service, not a browser grid.
  • You need formulas plus structural edits, undo/redo, persistence, and restore.
  • You are building a coding-agent or workflow-agent loop that must verify writes by reading formulas and values back from the same workbook model.
  • You want an MIT-licensed package surface.
  • You want benchmark claims tied to checked-in artifacts and local commands.

When Not To Choose bilig First

  • You need a mature commercial support channel today.
  • You need a mature engine with a longer compatibility history before adding reduced fixtures for the exact Excel edge cases your workflow uses.
  • You need a library already centered around a visual spreadsheet component.
  • You need every XLSX feature preserved across import/export right now.

For those cases, start with HyperFormula or a full spreadsheet product, then use bilig's compatibility notes to decide whether a narrower WorkPaper runtime fits a later slice.