openenvforge/3ambench
3ambench: Harbor dataset on Hugging Face with 75 tasks. 3amBench: can your agent write alerts that page the right human at 3 a.m., and only then?
Tasks
- slo-onboarding (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- slo-onboarding (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- slo-onboarding (medium): 9 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- slo-onboarding (medium): 9 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- slo-onboarding (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- slo-onboarding (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- alert-storm-cleanup (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- alert-storm-cleanup (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- alert-storm-cleanup (medium): 10 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- alert-storm-cleanup (medium): 10 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- alert-storm-cleanup (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- alert-storm-cleanup (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- missed-page-postmortem (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- missed-page-postmortem (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- missed-page-postmortem (medium): 10 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- missed-page-postmortem (medium): 10 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- missed-page-postmortem (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- missed-page-postmortem (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- latency-slo (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- latency-slo (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- latency-slo (medium): 9 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- latency-slo (medium): 9 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- latency-slo (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- latency-slo (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- team-reorg-migration (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- team-reorg-migration (easy): 5 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- team-reorg-migration (medium): 9 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- team-reorg-migration (medium): 9 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- team-reorg-migration (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- team-reorg-migration (hard): 14 change requests on a Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E1): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E2): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E3): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E4): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.
- expert (E5): a week of the observability queue on a large legacy Prometheus/Alertmanager repo, graded by promtool/amtool replays.