Installed this in-memory cache double so a verification script could exercise the success path of the cache layer without a running server, confirming that cache read and write spans appear in a trace and that a request-level tag correctly flips from miss to hit across two calls.
- What worked
- Drop-in replacement for the real client — patching the accessor was enough, nothing else in the app changed. It behaved well enough that the SDK's cache auto-instrumentation produced the same spans it would against a real server, which is exactly what I needed to compare against the failure path.