bilig

Launch Post: Headless Spreadsheet Engine For Agents

Status: ready-to-adapt public launch post for @bilig/headless.

Use this as the first broad technical post for Hacker News, Reddit, dev.to, a personal blog, or a project site. Adapt tone and length per channel, but keep the evidence and caveats intact.

For a Hacker News-specific submission checklist, title, first-comment draft, and preflight, use docs/show-hn-launch-pack.md.

Long-Form Version

Title:

A headless spreadsheet engine for AI agents and Node services

Post:

I built @bilig/headless, a TypeScript spreadsheet engine for agents, Node services, and local-first workbook automation.

The motivating problem is simple: a lot of useful business logic still lives in spreadsheet-shaped models, but most automation either drives a browser grid or reimplements formulas in ad hoc code. That is brittle for coding agents and awkward for backend services.

@bilig/headless exposes a WorkPaper API for programmatic spreadsheet work:

The repo includes a runnable external-consumer example:

git clone https://github.com/proompteng/bilig.git
cd bilig/examples/headless-workpaper
npm install
npm start

Or install the package directly:

npm install @bilig/headless

Minimal example:

import { WorkPaper } from "@bilig/headless";

const workbook = WorkPaper.buildFromSheets({
  Deals: [
    ["Region", "Customers", "ARPA", "Revenue"],
    ["West", 20, 1200, "=B2*C2"],
    ["East", 30, 250, "=B3*C3"],
  ],
  Summary: [
    ["Metric", "Value"],
    ["Total revenue", "=SUM(Deals!D2:D3)"],
  ],
});

const summary = workbook.getSheetId("Summary");
console.log(workbook.getCellValue({ sheet: summary, row: 1, col: 1 }));

Performance claims are checked in rather than left as marketing copy. The current WorkPaper benchmark artifact records 46/46 mean wins on scorecard-eligible comparable workloads against HyperFormula-style workloads: 38/38 public and 8/8 holdout.

That is not a blanket “faster on every p95 row” claim. The evidence note spells out what is measured, what is excluded, and where the p95 caveat still exists:

https://github.com/proompteng/bilig/blob/main/docs/headless-workpaper-benchmark-evidence.md

The project is still early infrastructure, not a finished Excel clone. Known open areas include deeper Excel formula parity, structured references, dynamic arrays, durable multiplayer sync, and more public release hardening.

Useful entry points:

Good first contributions are especially useful around formula parity fixtures, WorkPaper recipes, benchmark explanations, runnable examples, and docs that turn architecture notes into copy-pasteable code.

Hacker News Version

Title:

Show HN: bilig - a headless spreadsheet engine for agents and Node services

Text:

I built @bilig/headless, a TypeScript spreadsheet engine for programmatic workbook automation. It runs formulas, structural edits, persistence round trips, and validation without opening a browser grid.

The main use case is automation that needs spreadsheet semantics but should not screen-scrape Excel or Google Sheets. The repo has a runnable external-consumer example, npm package docs, and checked-in benchmark evidence against HyperFormula-style workloads.

It is early infrastructure, not a finished Excel clone. The current public claim is 46/46 mean wins on comparable WorkPaper benchmark workloads, with a documented p95 caveat.

Repo: https://github.com/proompteng/bilig

Reddit Short Version

I built @bilig/headless, a TypeScript spreadsheet engine for Node services and coding agents.

It gives you a WorkPaper API for formulas, structural edits, persistence, and range reads without opening a browser grid. The repo includes a runnable npm example and checked-in benchmark evidence against HyperFormula-style workloads.

Useful if you want spreadsheet-shaped business logic in code, or if an agent needs a real workbook API instead of screen scraping.

GitHub: https://github.com/proompteng/bilig npm: https://www.npmjs.com/package/@bilig/headless

Follow-Up Content Queue

  1. Why agents need workbook APIs, not spreadsheet screenshots.
  2. Persisting formula-backed WorkPaper documents in Node.
  3. What the WorkPaper-vs-HyperFormula benchmark does and does not prove.
  4. Adding formula parity fixtures without overclaiming Excel compatibility.
  5. Building a revenue model with @bilig/headless.