Morning Sentry error digest
Every weekday before standup, a prioritized digest of the last 24h across your Sentry projects — new issues, regressions, escalating problems, and the top issues by event volume.
Weekdays at 9:00 AMMonitoring Emailed to you
Use this routine →Opens prefilled — review and save.
Works with
The instructions
Every weekday at 9am, before standup, scan my Sentry projects for the last 24 hours and email me a prioritized error digest. Only read — never resolve, ignore, assign, or comment on any issue. 1. Scope the digest. List my projects (and org if needed) so the digest covers everything, or focus on production where that environment exists. 2. New issues. Search for issues first seen in the last 24h (is:new with firstSeen:-24h, level:error, environment:production where it applies). Capture each issue's short id, title, event count, affected user count, and permalink. 3. Regressed and escalating. Separately search is:regressed and is:escalating — these are problems that came back or are spiking beyond known noise. List them with the same fields. 4. Top issues by volume. Get the highest-volume issues over the last 24h (an event-count aggregation grouped by issue, sorted by count descending) so the loudest problems surface even if they're known. 5. Release context, where available. Pull recent releases and, only for projects that have Release Health set up, the per-release crash-free figures. Note error-volume trend versus the prior day. If a project has no session/health data, just skip the health line for it — don't invent a number. Write a skimmable email: a one-line headline leading with the single most important fact, then short sections (New, Regressed/Escalating, Top by volume, Release health where available). For each issue give the short id, title, counts, and a link. Keep it tight — surface the worst, don't dump everything. If a project, search, or release lookup returns nothing or errors, say so plainly for that section instead of guessing or padding. If there were no new or regressed issues at all, say it was a quiet night.
What a run emails you
Sentry digest — 30 Jun 2026, last 24h 1 new error spiking in checkout; payments regressed overnight. New (last 24h) • WEB-4821 — TypeError: cannot read 'total' of undefined · 1,204 events · 318 users — checkout step 2 [link] • API-372 — Timeout calling tax-service · 240 events · 96 users [link] Regressed / escalating • API-118 — PaymentDeclinedError (regressed) · 540 events · 210 users — back after release 2026.6.3 [link] • WORKER-55 — escalating, queue backlog growing · 1,890 events [link] Top by volume (24h) • WORKER-55 — 1,890 events • WEB-4821 — 1,204 events • API-118 — 540 events Release health • web 2026.6.3 — crash-free sessions 99.1% (down from 99.6%) • api — no Release Health data, skipped Quiet elsewhere — no new issues in mobile or marketing projects.
How it works
- Attach your Sentry connection — the routine enumerates your projects and reads issues, events, and releases for the last 24h, using only the access you granted. It only reads: it never resolves, ignores, assigns, or comments on an issue.
- The smart bit is prioritization: it separates genuinely new issues from regressed and escalating ones (things that came back or are spiking beyond known noise), then ranks the loudest problems by event volume — so the digest leads with what actually broke, not just what's noisy.
- Keep "Email me the result" on to get the prioritized digest in your inbox before standup; release-health figures appear per-release wherever Sentry's Release Health is set up.
Make it yours
- Narrow or widen the scope — focus on one or two critical projects, pin it to your `production` environment, or raise the bar to only surface issues above a user-count or event-count threshold.
- Adjust the window or cadence — digest the last 12h for a tighter morning view, or run it twice a day during an incident-heavy week.
- Reframe the sections — lead with regressions, group by project or team, or fold in recent releases and their authors for quicker triage.
Your connection stays yours: tokens are encrypted at rest, and the agent only uses the access you grant — and this routine only ever reads.
Run “Morning Sentry error digest” on autopilot
Copy it to your account, tweak the details, and it runs weekdays at 9:00 am.
Use this routine →More Monitoring templates
A
AI Share-of-Voice weekly monitorMonitoring
Every Monday, run a set of buyer-intent searches and tally which tools show up when people research your space — a weekly Share-of-Voice report, emailed.
Mondays at 8:00 AM
Daily web-scrape digestMonitoring
Every weekday, run an Apify scraper you choose, pull the fresh results, and email a summarized digest of what's new.
Weekdays at 8:00 AM
Daily brand mentions digestMonitoring
Every morning, get the X/Twitter mentions worth replying to, grouped by sentiment.
Daily at 8:00 AM
D
Daily competitor pricing watchMonitoring
Get an email the moment a competitor changes their pricing page — plans, prices, or limits.
Daily at 8:00 AM