Skip to content
Back to skills

Inbox Zero

ASecurity

Summarizes your daily.dev notifications in one short briefing, grouped by what needs you (replies, mentions, squad reviews) versus what's just news, and marks them read when you say so.

  • 8 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added October 17, 2026
ai-agentsgorailsapi

Works with

  • cursor
  • api
  • mcp

Security analysis

A100/100

Scanned October 17, 2026

npx -y skills add dailydotdev/marketplace --skill inbox-zero --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Inbox Zero?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Inbox Zero
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/dailydotdev-inbox-zero/badge)](https://www.skillsdirectory.com/skills/dailydotdev-inbox-zero)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: inbox-zero
description: Summarizes your daily.dev notifications in one short briefing, grouped by what needs you (replies, mentions, squad reviews) versus what's just news, and marks them read when you say so.
---

# Inbox Zero

Notifications pile up faster than you read them. Inbox Zero turns the backlog into one briefing and clears it on your command.

## Requirements

One of:

- The daily.dev MCP server (`https://api.daily.dev/mcp`) connected to your agent (add the `write` scope to mark notifications read), or
- A daily.dev personal access token in `DAILY_DEV_TOKEN` (https://daily.dev/settings/api). Base URL `https://api.daily.dev/public/v1`, header `Authorization: Bearer $DAILY_DEV_TOKEN`. Never send the token anywhere except `api.daily.dev`.

Each step lists the MCP tool first and the REST call after it.

## Steps

1. **Count first.** `getNotificationsUnreadCount` (`GET /notifications/unread/count`). If it's zero, say so and stop.
2. **Fetch.** `getNotifications` with `limit` 50 (`GET /notifications?limit=50`); page with `cursor` while `readAt` is null and the count isn't covered yet. Each notification has `type`, `title` (may contain HTML tags, strip them), `description`, `targetUrl`, `avatars` and `attachments`.
3. **Triage by `type`** into three buckets:
   - **Needs you**: replies and mentions (`comment_reply`, `comment_mention`, `post_mention`), squad moderation (`source_post_submitted`, member requests), anything addressed to you personally.
   - **Social**: upvotes, awards, milestones, new followers.
   - **Content**: new posts in sources and squads you follow (`source_post_added`), agent and briefing digests (`interest_content_batch`), announcements.
   Unknown types go to **Other** with their raw `type` shown.
4. **Write the briefing.** Needs-you items one per line with a link (`targetUrl`) and who it's from. Social collapsed into one sentence with counts. Content as a short list of the most relevant titles (use `attachments[].title` when present), max 8, link each. Lead with the needs-you count.
5. **Offer to clear.** Ask: "Mark all N as read?" Only on a yes, call `postNotificationsRead` (`POST /notifications/read`). It marks every notification read; there is no per-item variant, so never call it before the user has seen the needs-you list.

## Output

Under 200 words for a typical backlog. Three headers, links on everything actionable, then the clear prompt.

## Guardrails

- Never mark read without an explicit yes in the same conversation.
- Don't act on notifications (no replying, approving or following); link them so the user can.
- Two or three requests to read, one to clear.

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in ai-agents →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.