n8n alerts where your team actually lives.

What ships

In each workspace's alert settings you paste a Slack incoming webhook URL (or any https endpoint that accepts JSON). Every alert that emails also POSTs a Slack-blocks payload to that URL. A "Send test" button confirms reachability before you save.

The payload uses Slack's blocks format: a header with the workflow name, a section of labeled fields (workspace, failure mode, detail, transition), and an "Open in WorkflowRadar" button that deep-links into the dashboard. Discord accepts the same shape with ?wait=true appended to the webhook URL.

Defense in depth

• SSRF guard: webhook URLs must be https on a public hostname. Localhost, RFC1918 private ranges, link-local, and the AWS metadata endpoint are refused before the fetch is attempted. Redirects are not followed (no 30x → internal hostname trick).
• 5-second timeout: a slow webhook can't block a sync. A single retry on 5xx covers Slack's occasional 503 under load.
• 10 KB payload cap: a pathological alert payload can't blow the webhook or the audit row.
• Audit log: every dispatch attempt writes a row to webhook_dispatch_attempts with status code, error message, and payload size. Append-only.

Smart alert rules

The same panel where you wire up the webhook URL exposes two rule-based alert types that fire independently of n8n's status field:

• Consecutive failures: alert when a workflow fails N runs in a row (2–50). Independent of the workspace health transitions, so a workflow that's already "critical" still alerts you again when it hits a new threshold.
• Error-rate threshold: alert when a workflow's 30-day error rate crosses a percentage you configure (1–100%). Catches workflows that fail sporadically but at a meaningful clip — the kind that don't trip the consecutive-failure rule but quietly burn down customer trust.

Each rule has its own dedup bucket. Toggling one off doesn't affect the others. The toggle to disable health-status alerts doesn't disable the rules.

Cooldown batching

When a workspace gets noisy — say a cron broke and is firing every 15 minutes — we batch instead of flooding. After the third alert of the same type within 15 minutes the next post gets a context footer: "+5 more failure alerts in the last 15 min." The first three posts still go through normally so a real incident gets seen.

Setup

One field, one button. In your workspace's alert settings:

1. Create a Slack incoming webhook for the channel you want (Slack docs →).
2. Paste the URL into the "Slack / webhook URL" field.
3. Click "Send test" — you should see a test message in your channel within a couple of seconds.
4. Save. From now on every alert routes to both email and Slack.

Available on

Slack and webhook delivery is on every paid tier — same as email alerts. No upcharge. The 60-second setup is the same on Consultant and Agency. See pricing.