Build, audit and grade Excel model workbooks: layout standard, dependency-free command-line auditor (errors, hard-coded numbers, broken patterns, circular references, controls, stale values) and rubric-based grading aid.
Installing only copies or configures. Nothing is executed for you. The Skills CLI verifies the checksum before writing files.
An Agent Skill for building, auditing and grading Excel model workbooks.
.xlsx / .xlsm files without Excel
and reports errors (root causes first), circular references, hard-coded numbers, broken formula
patterns, failing or fake controls and stale stored values; it also traces precedents and
dependents across sheets.skills install @spboucher/excel-model-workbook --client claude-code
or copy the folder into your agent's skills directory (for Claude Code: ~/.claude/skills/).
node scripts/xlsx.mjs --help
node scripts/xlsx.mjs report model.xlsx
node scripts/xlsx.mjs grade reference.xlsx student.xlsx --rubric assets/rubric.example.json
Node.js 20 or later, no npm install. Nothing in a workbook is executed; the tool writes only
the report or evidence file you ask for, and never overwrites.
| Path | What |
|---|---|
SKILL.md | Instructions for the agent |
scripts/xlsx.mjs | Command-line auditor, 14 commands, JSON output |
scripts/lib/ | Engine, identical to the one in @spboucher/xlsx-audit-mcp |
references/modeling-standard.md | Layout standard and an openpyxl skeleton |
references/audit-procedure.md | How to read each result |
references/grading.md | Rubric format and grading policies |
references/defects.md | Defect catalogue |
assets/rubric.example.json | Rubric template |
.xlsx and .xlsm only. The built-in evaluator covers arithmetic, comparisons, SUM,
AVERAGE, MIN, MAX, ROUND, IF, references and plain defined names; every other formula
is read from the result stored in the file, and the tool states the share it could not recompute.
Macros, external links, charts, pivot tables, conditional formatting and data validation are not
read. A clean report shows that mechanical defects are absent, not that the model is right.
The same engine is available as an MCP server: @spboucher/xlsx-audit-mcp.
MIT
Select a file to preview it. Files are shown as plain text and never executed.
No advisory findings from static-1.0.0 on Nov 10, 2026. 9 informational notes.
Static checks are advisory. They never certify that a package is safe. Compatibility entries are publisher claims unless marked verified.
| Version | Published | Size | SHA-256 | State |
|---|---|---|---|---|
| 1.0.0latest | Nov 10, 2026 | 47 kB | ce5ca08c36ee… | published |
Run a sequential revision experiment on a text and record what the Pangram detector reports for each version, as an observed outcome: frozen editorial plan, one submission per step, real evaluations, preservation checks, chart and report.
Peer-review protocol for LaTeX and Markdown manuscripts: deterministic audit script, five review roles, adversarial verification of findings, anchored rubric, referee report and decision letter templates.
Verify citations against Crossref, OpenAlex, arXiv and doi.org: a DOI must resolve and its metadata must match. Audits .bib files, resolves references without DOI, flags retractions, guides literature searches.