Triage Digest Template

Module 5, Lesson 5.2 · produces Entry 1 of the capstone inbox-and-calendar log · concept shape + recipe run

What this is. The six-category digest shape the agent is directed to produce against your bounded agent-access label, plus the four-step audit you run against it afterwards.

How to use. Hand the agent this shape inside your triage prompt. When the digest returns, run the audit before you act on anything.

Safety norms — copy verbatim to the top of every digest.

Read-only. No drafts, no archive, no marking read. Drafts, not sends. Least access for the task.

Header

Student
Digest date / time window
Tool used (Cowork-tab session, or Claude Code CLI working directory)
Scope (label name + message count)

The six categories

01 — Reply today
Messages that need a reply from me in the next 24 hours. Agent explains why in one phrase per message.
02 — Reply this week
Messages that need a reply, but not today.
03 — FYIs
Messages I should know about but do not need to act on. One line each.
04 — Archivable
Newsletters, receipts, auto-mail. Agent lists them so I can confirm nothing mis-classed.
05 — Flagged as unclear (honesty category)
Messages the agent does not have enough context to classify. This is a feature; do not pressure the agent to empty it.
06 — Missed by the fence (priority category)
Messages the agent thinks probably belong in the agent-access label but are currently outside it. Agent does not move them; it just notes them.

The four-step audit (yours, not the agent’s)

Count

Does the total number of messages in all six categories add up to the number of messages in my label?

Placement

Pick three messages at random from the digest. Does each one actually belong in the category the agent put it in?

Invention

Does any line in the digest describe a message I do not recognize, or a sender I never saw in my label?

Missing context

Does the “flagged as unclear” category exist and have at least one message in it, or has the agent over-claimed? Empty unclear buckets are usually a warning, not a win.

Closeout

All six categories present; every message in the label accounted for.
All four audit checks performed, with a one-line note each.
Saved as digests/<YYYY-MM-DD>.md in the capstone log.
Safety norms visible at the bottom of the entry.

A note on the audit

The audit is the reason the digest is safe to trust. Skipping it because “this one looks fine” is a common Module 5 failure. The whole point of the four-step audit is that looking fine is exactly what a subtle digest miss looks like. Run all four checks every time, for at least the first two weeks. After that, run at least two of them every time, forever.

Print this page. Use it.