# Inbox Triage

> >

- **Type:** Skill
- **Install:** `agentstack add skill-ootto-ai-claude-inbox-skills-inbox-triage`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [Ootto-AI](https://agentstack.voostack.com/s/ootto-ai)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [Ootto-AI](https://github.com/Ootto-AI)
- **Source:** https://github.com/Ootto-AI/claude-inbox-skills/tree/main/skills/inbox-triage

## Install

```sh
agentstack add skill-ootto-ai-claude-inbox-skills-inbox-triage
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Inbox Triage

Sort a full inbox into reply-now / later / ignore, with drafts for the urgent ones.

## When to use
Your inbox is full and you need to know what actually needs a response today, what can wait, and what to archive — plus ready-to-send drafts for the urgent ones.

## What you'll need
A list of unread emails (sender, subject, and a line or two of each message — paste as many as you want).

## Instructions
Collect any missing inputs from the user, then run this prompt:

```
You are my executive assistant triaging my inbox. Below is a list of unread emails, each with sender, subject, and a short preview.

[paste your emails here — one per block, with sender, subject, and preview]

Sort every email into exactly one of three buckets:
1. REPLY NOW — needs a response today (a real person waiting, a deadline, money, or a decision blocked on me)
2. LATER — matters but can wait this week (FYIs I should read, low-urgency requests)
3. IGNORE — newsletters, promos, automated notifications, or anything I can safely archive

For each email give one short line: sender — why it's in that bucket.
Then, for every REPLY NOW email only, write a ready-to-send draft reply: plain, friendly, professional, 2-4 sentences, no filler. If a reply needs information only I have, leave a clearly marked [blank] for me to fill.

Output as three headed sections (Reply Now, Later, Ignore), with the drafts under each Reply Now item.
```

**Tip:** Paste the actual preview text, not just subject lines — the more of each message Claude can see, the better it judges urgency and writes a usable draft.

---

Built by **[Ootto](https://www.ootto.ai)** — the AI autopilot that connects your tools once and runs invoicing, follow-up, and reports for you, automatically. [Book a demo →](https://www.ootto.ai)

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [Ootto-AI](https://github.com/Ootto-AI)
- **Source:** [Ootto-AI/claude-inbox-skills](https://github.com/Ootto-AI/claude-inbox-skills)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-ootto-ai-claude-inbox-skills-inbox-triage
- Seller: https://agentstack.voostack.com/s/ootto-ai
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.