skills/bilig-workpaper/SKILL.md
Formula WorkPaper runtime and MCP server for AI agents and Node.js services. Use when an agent needs spreadsheet-style formulas, cell edits, recalculation, readback verification, or persisted WorkPaper JSON without driving Excel UI.
npx skillsauth add legendaryabhi/agent-skills-hub bilig-workpaperInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Bilig WorkPaper gives agents a reviewable spreadsheet model without requiring browser or Excel UI automation. Use it to edit cells through an API, recalculate formulas, verify display-value readback, and persist WorkPaper JSON for audit or restore flows.
Use the package directly with npm exec when you only need the MCP server:
npm exec --package @bilig/headless -- bilig-workpaper-mcp --workpaper ./pricing.workpaper.json --init-demo-workpaper --writable
For TypeScript code, install the package in the project:
npm install @bilig/headless
Keep workbook state in a WorkPaper JSON document. Write inputs through the WorkPaper API or MCP tools, recalculate, then read the computed display value back from the model.
A file-backed MCP session should expose tools for listing sheets, reading cells or ranges, setting cell contents, validating formulas, exporting the document, and reading display values after recalculation.
After each write, read the changed input cell and the dependent output cell. Keep the exported WorkPaper JSON when the task needs proof, rollback, or a reproducible bug report.
npm exec --package @bilig/headless -- bilig-workpaper-mcp --workpaper ./pricing.workpaper.json --init-demo-workpaper --writable
An agent can then update Inputs!B3, read Summary!B3, and export the updated WorkPaper JSON.
import { WorkPaper } from "@bilig/headless";
const paper = new WorkPaper();
paper.setCellContents("Inputs", "B2", 1200);
paper.setCellContents("Inputs", "B3", "=B2*1.2");
paper.recalculate();
const result = paper.getCellDisplayValue("Inputs", "B3");
console.log(result);
Problem: The agent wrote a formula but the output did not change.
Solution: Recalculate the WorkPaper, then read both the input cell and the dependent display-value cell.
Problem: The workflow needs to preserve state across agent turns.
Solution: Run the MCP server with --workpaper ./file.workpaper.json --writable and export the document after important changes.
development
Multiagent AI system for scientific research assistance that automates research workflows from data analysis to publication. This skill should be used when generating research ideas from datasets, developing research methodologies, executing computational experiments, performing literature searches, or generating publication-ready papers in LaTeX format. Supports end-to-end research pipelines with customizable agent orchestration.
development
Implement DeFi protocols with production-ready templates for staking, AMMs, governance, and lending systems. Use when building decentralized finance applications or smart contract protocols.
tools
NGS analysis toolkit. BAM to bigWig conversion, QC (correlation, PCA, fingerprints), heatmaps/profiles (TSS, peaks), for ChIP-seq, RNA-seq, ATAC-seq visualization.
testing
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.