/master-schedule

What it does

Type /as:master-schedule at the start of a project and Claude initializes the product-library.csv that every other materials skill reads and writes. Run it once — the rest of the product workflow depends on it.

The library is a plain CSV file that lives in your project folder. Nothing leaves your machine: no account, no connector, no spreadsheet service. The skill finds your project root, previews the exact path before writing anything, and creates the 33-column header on your confirmation.

It also validates. Point it at an existing library and it reports the row count, or stops and explains the header or row error without touching the file.

Install

Claude Desktop:

1. Open Customize → Browse plugins
2. Click + → Add marketplace from GitHub
3. Enter AlpacaLabsLLC/skills-for-architects
4. Install Architecture Studio

Claude Code (terminal):

claude plugin marketplace add AlpacaLabsLLC/skills-for-architects
claude plugin install as@skills-for-architects

Or symlink just this skill:

git clone https://github.com/AlpacaLabsLLC/skills-for-architects.git
ln -s $(pwd)/skills-for-architects/skills/master-schedule ~/.claude/skills/master-schedule

Usage

/as:master-schedule

Run from anywhere inside your project. The skill walks up the directory tree to the nearest folder containing a PROJECT.md and treats that as the project root — the library is created there, never outside a recognized project.

The 33-column schema

product-library.csv uses one row per product, organized by section:

Section	Columns	What’s there
Product Identity	A–K	Category, brand, vendor, name, designer, SKU, link, collection
Dimensions	L–Q	W, D, H, seat height, unit, weight
Materials & Finish	R–T	Materials, colors/finishes, selected finish
Pricing	U–W	List price, sale price, currency
Logistics	X–AA	Lead time, warranty, certifications, COM/COL
Meta	AB–AC	Clipped-at timestamp, image URL
Research	AD–AG	Tags, notes, status, source

Category uses one of 22 canonical terms, so rows stay sortable across skills. Dimensions are numeric with the unit in its own column — the manufacturer’s original unit, never silently converted. Empty means unavailable; no N/A or — placeholders.

The file is UTF-8 without a BOM, RFC 4180-quoted, with a deterministic column order. Writes are validated in full and replaced atomically, so a failed operation leaves the file byte-for-byte unchanged.

Projects that track embodied carbon get an optional epd-library.csv alongside it, on a 42-column schema.

If you already have product data

Import is always explicit — the skill never sweeps up a stray CSV on its own. Export what you have, then point the skill at it:

/as:master-schedule

Then ask to import your file. The skill states the source and target paths, validates the entire source against the exact 33-column schema, and asks once before writing. It refuses a malformed source, and it refuses to overwrite a library that already has rows.

Upgrading from a connected sheet

Architecture Studio removed spreadsheet connectivity in 1.4. If your project has a master-schedule.json or canoa.json from an earlier release, those files hold configuration — not recoverable rows.

The skill leaves them exactly where they are, byte-for-byte. To bring your data across: export the former sheet as CSV yourself, then run the explicit import above. Nothing is fetched on your behalf and no rows are reconstructed from the JSON.

Pairs with

Every product skill in this plugin reads from the library this skill creates:

• /as:product-research — find products from a design brief
• /as:product-spec-bulk-fetch — extract specs from product page URLs
• /as:product-spec-pdf-parser — extract specs from PDF catalogs
• /as:product-data-import — bring an existing product list into the schedule
• /as:product-enrich — auto-tag categories, colors, and materials