~/home / n8n-live
Automation live.
A robot writes this page. An n8n workflow on my server queries seven sources, an AI sums it all up in two sentences, and the result lands here. It runs by itself every 6 hours, or right now if you press the button.
The robot is warming up …
Where is the ISS? And where did the earth shake?
The pulsing dot: the ISS at the last run. Orange: the 15 strongest quakes of the last 24 hours (magnitude 2.5+).
Recent runs
Each bar is one run; height = run time.
What happens here
A webhook (your click) or a 6-hour schedule starts the workflow. Seven HTTP nodes query public APIs in parallel, a Merge node waits for all of them, a Code node bundles the data, a language model (gpt-oss-120b at IONOS, in Germany) writes the bulletin, a second Code node checks the answer, and a final HTTP node sends everything to this page. If one source fails, the rest still runs.
How the workflow is built: n8n introduced →
What else runs automatically →
A new workflow to rebuild every Thursday →