# Regression corpus for the availability safety-simulator campaign.
#
# One u64 seed per line; `#` comments and blank lines are ignored. Every seed here runs on every
# pull request across the full topology matrix, so a break the scheduled campaign found never
# reappears unnoticed. When a scheduled shard reports a violating seed, add it here (with the topology
# and the fix in the same change) to promote it into the permanent pull-request guard.
#
# Empty today: the model holds every invariant, so no seed has failed. The scheduled campaign sweeps a
# large seed space to find the first one.
