Answer Radar

The Email Brief and Its Preferences

The brief is the recurring decision for your account: what moved across every active project since the last window, which queued actions are outstanding and which are new, and what was completed or dismissed. It is off until you turn it on, it runs on the cadence and timezone you choose, and it is honest enough to tell you when nothing meaningful changed.

Quick Answer

On the Answer Radar page, the Email briefs panel has a Turn on briefs button and cadence and timezone controls. Weekly is the default. Each brief compares the last window with the equal window before it across active projects, lists outstanding and new actions and closed ones, and returns nothing_meaningful_changed with the reason when all three are quiet.

  • Off by default for every account. Only your own toggle enables it.
  • A window that cannot be compared is reported as not comparable, never as a fall to zero.
  • Every number carries its disclosure; there is no confidence value and no written summary.

Turning it on

Open /answer-radar. Below the project list, the Email briefs panel shows the current state. Press Turn on briefs, choose a cadence and a timezone, and the next slot is computed in your local hour, preserved across daylight-saving changes by the same cadence rule the audit scheduler uses. Turn off briefs stops them. Nothing else in the product enables a brief for you; there is deliberately no 'enable for everyone' path.

Every brief email carries an unsubscribe link, and sending is additionally gated by an operator switch that is off unless the operator has turned it on, so no account is mailed without two deliberate acts.

What a brief contains

Three things, per the answer_digest_get tool the brief is built from. Movement: what changed across every active project in the last N days versus the equal window before it, per segment, with each number's disclosure. Actions: which queued cards are outstanding, which are new since the last brief, and which were completed or dismissed. And a verdict when all three are quiet: nothing_meaningful_changed, with the reason.

A window that could not be compared, because a run failed or an instrument changed, is reported as not_comparable, never as a fall to zero. No confidence value is reported and no summary sentence is written. The brief is the evidence for the week; the sentence about what to do with it is yours, or your assistant's if you read the digest over MCP.

How duplicates are prevented

The slot is a claim, not a log. Before an email is sent, the next slot is advanced on the row it read, and only the run that wins that write sends. Two concurrent runs of the sender cannot both mail you. The cost is that a crash between claim and send loses one brief, which is the right direction to fail for email: a missing brief is forgiven far more readily than a duplicate one.

  • Weekly is the default cadence. Monthly is available for quieter accounts.
  • Set the timezone you read in; the brief lands in that local hour.
  • Read the same data any time with answer_digest_get, with your own days and cadence.

FAQ

Why did I get a brief saying nothing changed?

Because the movement across projects stayed under what the measurement can resolve, no actions were added or closed, and nothing was completed or dismissed. The brief says so with the reason rather than inventing a highlight.

Can the brief be sent to my client?

The brief goes to the account's email. For a client, create a share link to the project's metrics, which shows the summary without a seat.

Does turning on briefs start audits?

No. The brief reads what audits, scheduled or manual, have already measured. Schedules are set separately per project.

Related help pages