The Problem
Every engineering team eventually hits the same wall. Your monitors fire, your alerting pipeline dumps raw JSON into Slack, and someone on-call has to read through a wall of text at 3 AM to figure out what broke, which service is affected, and what ticket to open.
It's manual. It's slow. It's toil.
What NKAI Gateway Does
NKAI Gateway sits between your monitoring tools and your issue trackers. It:
- Ingests webhooks from any source — Prometheus, Grafana, Datadog, PagerDuty, or a custom HTTP endpoint
- Routes events declaratively based on rules you define in a simple config file
- Transforms payloads with a template engine, so your tickets look exactly how your team expects them
- Provisions issues in Jira, Linear, GitHub Issues, or any other tracker automatically
No more copy-pasting alert details into tickets. No more lost context between your monitoring and your issue tracker.
How It Works
routes:
- match:
source: prometheus
severity: critical
transform:
title: "[{{ severity | upper }}] {{ alert_name }} on {{ instance }}"
body: |
**Service:** {{ labels.service }}
**Started at:** {{ starts_at }}
**Description:** {{ annotations.description }}
destination:
type: linear
team: infrastructure
priority: urgent
Define a route. Match on any field in the incoming webhook. Transform the payload using Jinja-style templates. Route to your destination. That's it.
Self-Hosted and Open Source
NKAI Gateway is designed to run on your infrastructure. A single docker-compose up -d gets you running in minutes.
docker-compose up -d
No data leaves your environment. No vendor lock-in. Full control over your alerting pipeline.
What's Next
We're working on:
- A web UI for managing routes visually
- More destination integrations (Notion, Asana, Shortcut)
- An AI enrichment layer that classifies alerts and suggests fixes before the ticket is even created
Follow along as we build.