Properties Instead Of A Brief

A task with a perfect set of fields and an empty body is not a task. It is a reminder that something exists.

Pattern

An agent creates tasks in a tracker. The fields are perfect: owner, due date, priority, area, tags, links. The body is empty, or holds a fragment like follow up re pricing.

Everything a machine needs is present. Everything a human needs is not.

Why it looks right

The metadata is the part that is hard to get right and easy to verify, so filling it well feels like the job being done. A populated task also looks complete in a list view, which is where anyone reviewing the system will see it.

And at the moment of creation the context is obvious — the conversation just happened.

Why it fails

Context evaporates on a schedule of about 3 weeks. After that, follow up re pricing is a message from a stranger. The owner either guesses, asks, or quietly leaves it — and the third outcome is the common one, because a task nobody understands is easier to postpone than to interrogate.

The metadata cannot help. Tags describe where a task belongs, not what it is.

Instead

Require a body with a fixed shape, and treat an empty one as a defect rather than a style choice:

What this is about — 2 to 4 full sentences, abbreviations spelled out

What needs doing — actionable checklist

Sources — links to the evidence

The test is a single question, applied before saving: would this be understood by someone with no memory of the conversation that created it?

Two supports make it stick. The agent writes the body first, then the fields, so the reasoning is not squeezed in as an afterthought. And an empty body fails a check rather than producing a slightly worse task — otherwise it will be skipped exactly on the busy days when the task matters most.