# Inbox Triage

> Sort an inbox into what needs a reply, what needs a decision, and what is just noise. Use when the user asks to triage, catch up on, clean up, or make sense of their inbox.

- Skill: `amd/inbox-triage-2` (Agent Skill)
- Install (CLI): `npx get-skill add amd/inbox-triage-2`
- Raw SKILL.md: https://api.skillmd.com/api/skills/amd/inbox-triage-2/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity
- Author: amd (https://skillmd.com/u/amd)
- Updated: 2026-11-16
- Page: https://skillmd.com/skills/amd/inbox-triage-2

---


# Inbox Triage

Triage answers one question per message: **does a human have to act on this?**

- Scan first (`pre_scan_inbox`, then `triage_inbox`); never open messages one by
  one to build a picture.
- Buckets, in order: needs a reply · needs a decision · for information · noise.
- Open only what the category can't settle. Sender + subject usually decide it.
- Act on the reversible end only — mark read, star, label, archive. Reply, send,
  and delete are proposals the user confirms.

**Report** one opening sentence and stop: how many items need attention, and
how much of the mailbox was scanned. The numbered breakdown is rendered from
the scan itself — do not write it out, and never re-list, renumber, or
summarise those items yourself.

Age beats volume. A thread the user already answered is handled. "URGENT" in a
subject line is a claim, not a fact.

