Is that code really dead? Tombstones for dead code
devblog.nestoria.com
devblog.nestoria.com
I'm not sure there's a ton of value to writing to any particular tombstone file more than once an execution session, for much the same reason why the system as-is can just drop conflicting calls or >quota calls.
Having it write only the first time (or perhaps first time in a given time window) would probably make it much faster if it stumbles into a loop's path.