new·The score now tells you which way it movedA brain's exam only ever grows: its own material writes questions, and so does every question a real caller asked and did not get answered. The score is a percentage over that growing set, so a brain that learned more could post a smaller number — and this week three did. One of them answered two MORE questions than the week before and showed eighteen points less. Printed as a single percentage, that reads as decline to a reader and as punishment to anyone who contributes material.all news →
mozg.beta
Sign in

Building With AI Agents · all subjects

deploy/build

2 notes, read out of this brain and free to use. Each one was extracted from a source and is re-checked against its exam.

Build without credentials in Docker, validate at runtime

Next evaluates every module while collecting page data, and lib/env refuses an empty environment by design — so the moment a layout imported it for a base URL, the container build died. The build stage gets placeholders. They live only in that layer; the container reads the real environment at runtime and validates it there, which is the moment that actually matters. Without this, any future import of lib/env from an always-evaluated module breaks the deploy.

Do not collect opengraph at build time

The opengraph route reads the brain from the database, so Next tried to collect its page data, lib/env refused to parse an empty environment, and the deploy failed. Build stages should not require database access.

Give your agent this brain