STORE PACKAGE
AGPL-3.0 · SELF-HOSTED · YOUR KEYS

Engineering · Store package · v1.1.0

AI Bake-Off

Race one job across every AI lane you already have, grade every output on the shared quality judge, and get the cheapest lane that still clears your quality bar.

Opens ?app=bake-off on the demo deployment. On your own box it is the same URL.

2 screens in the ribbon · opens on bake-off-home

AI Bake-Off ReviewAI Bake-Off

How it runs

What happens when you use it.

Not a chat window with a plugin behind it. Every step below is something the app declares in its manifest, which is why it can be told to you plainly.

  1. 01

    You open it, or the assistant does

    Opening ?app=bake-off loads its 2 screens into the shared cockpit, starting on bake-off-home. The ribbon, the ticket list and the assistant are already there — the app contributes views, not a second application to log into.

  2. 02

    Or you just ask

    The assistant classifies what you asked for and hands it to the bot that owns that domain. You do not pick the app, and you do not learn a command — routing is by declared capability, not by keyword matching a menu.

  3. 03

    A named bot picks it up

    bake-off-analyst is accountable for the work. 3 declared capabilities say what it is trusted to do; the tokens and dollars it spends are recorded against that identity, so the cost has a line item rather than disappearing into a monthly bill.

  4. 04

    Longer work becomes a ticket

    Scheduled and swarm-initiated work enters the queue as a bake-off ticket. It is durable, it has an owner, you can watch it move through its phases, and if it fails it fails somewhere you can open and read.

  5. 05

    The result lands where you can check it

    Back on the app's own screens, as state you own — not as a chat message you have to scroll back to find.

With the assistant

You can ignore this app entirely and just ask.

The assistant is the front door. It classifies your request, delegates to the bot that owns the domain, and answers in one voice — so you do not have to know which of the applications this happens to live in.

Accountability

3 capabilities, each attached to a named identity.

Declared in the manifest and bound to the bot above — so "what is this thing allowed to do" is a list you can read, not a prompt you have to trust.

cost-quality-analysismodel-selection-adviceevidence-summarization

What it needs

Before it can do anything for you.

Also on the engineering shelf

Apps that pair with this one.

All 12 engineering apps →

Underneath

What every app on this site is standing on.

This app does not ship its own queue, its own credential store or its own cockpit. It declares what it needs and the platform provides it.