Advisory 07 — twelve more avatars, made and waiting. We think you should take them.

To: docade · From: WootBuild · Date: 2026-08-25 Status: OPEN — recommendation, yours to accept or decline Trigger: finishing the twelve avatars your manifest asks for.

The short version

Your twelve avatars are delivered. We then made twelve more you did not ask for, in the same style, and we would like you to ship them.

They are not in assets/avatars/ and they will not be until you say so. Your bridge is unambiguous —

WootBuild consumes it and never invents a row — a piece that is not in it does not exist.

— so twelve pieces with no rows have no home, and writing them into your asset tree anyway would break the one rule that document states outright. The work is finished and parked. Adding the rows is a one-line-each change on your side and we are not making it for you.

They cost you nothing. No invoice, no budget line, no obligation. Decline and nothing happens.

What they are

Twelve creatures, chosen to be the ones a child actually wants to be:

Every one is smiling — an open, obvious smile that survives 48px — and every one takes baby proportions: oversized head, large low-set eyes, blunt shapes where the real creature would have points. The traditionally frightening ones are friendly: the yeti is delighted, the chupacabra is approachable, bigfoot is pleased to see you.

The rule we held them to, because it is the one that makes a set like this fail: cute is a treatment, not a replacement. A chupacabra that reads as a generic puppy has failed. A jackalope without antlers is only a rabbit.

Why we think you want them

Twelve is a thin roster for an identity choice. A child picks an avatar once and looks at it every day. With twelve options, a household of three siblings has already used a quarter of the set, and the four-and-under end of your audience is choosing between animals they may have no feeling about.

These are the ones with pull. A unicorn and a dragon are not the same kind of choice as a turtle. Neither is an ice cream sundae with a face, which is the piece we would bet on and the one we would have been least likely to make if we were only filling a row in a spreadsheet.

They cost you nothing to hold. They are SVG, they are the same 256px contract as your existing twelve, and they sit at roughly 2 KB each. Shipping all twenty-four costs you a picker that scrolls.

What it takes on your side

Twenty-four is the number we would ship. If you want them:

1. Add twelve rows to assets/asset-list.csv, category avatars, Size (px) 256, Format svg, paths following your existing convention — assets/avatars/unicorn.svg and so on.
2. Regenerate assets/manifest.json with npm run assets:csv.
3. Tell us, or just push it — art intake docade reads your work-list every session and will see the new rows.
4. We deliver them into the paths you declared, through the normal gate, with a receipt.

If you want fewer than twelve, add only the rows you want. We will deliver exactly what your work-list asks for and hold the rest. If you want none, say so and we will retire them; nothing is blocked either way.

What we are NOT asking

• Not asking you to change your manifest schema.
• Not asking for a decision this week. The work is done and it keeps.
• Not asking you to take all twelve. The sundae is the risky one and it is fine to leave it.

The one thing worth checking first

Look at all twenty-four at 48px in one row, which is how a child meets them. That is the only view where a set like this fails, and it fails by two pieces reading as the same thing rather than by any one piece being bad. We have checked it and are satisfied — narwhal against your existing whale was the collision we watched for — but it is your product and your call.

Ours to hold: docade-avatar-plus@1, twelve pieces, human-verdicted and frozen. Say the word and they are yours.