Scenarios overview
The other "build on top" pages cover one axis at a time — a custom drain, a custom enricher, a stream server. The scenarios in this section show how to combine them to solve specific real-world jobs.
Each scenario is opinionated, copy-paste oriented, and self-contained: drop the code in your project and it works.
Scenarios
| What you get | Axes used | |
|---|---|---|
| Local debugging toolkit | A live event panel + replay-from-disk in any app | observers/stream-server + observers/fs-reader |
| Tenant-aware logging | Every event carries the right tenant id, automatically | pipeline/custom-enrichers + pipeline/plugins |
| Compliance audit | A tamper-evident audit pipeline shipping to your secure backend | sinks/custom-drains + audit catalog |
| Cross-app error vocabulary | One @my-org/evlog-errors package every service depends on | shared/catalogs-as-packages |
When you should use this section
Use the scenarios when you know roughly what you want to build but not which evlog primitives to combine. Each scenario links back to the canonical pages for each surface, so you can drill down once you've found the recipe that fits.
Tail sampling
Decide post-hoc whether to keep an event based on the request outcome — keep all errors, keep slow requests, keep specific paths, drop the rest.
Local debugging toolkit
A live event panel + replay-from-disk you can drop into any app — combines the stream server, the fs reader, and a tiny consumer page.