Use $simulation-helper-skill-for-eclipse-sumo or Torii to audit one of the bundled SUMO/TraCI signal-control cases before reporting a performance claim.
fixed_time/: fixed-time baseline readiness.max_pressure/: pressure calculation and paired comparison readiness.data_informed/: causal information use, ablations, and metric evidence.
- What is the exact controller identity and what evidence proves it ran?
- Are routes, demand, seeds, horizon, outputs, TLS mapping, and completion criteria paired fairly?
- Are unfinished vehicles, teleports, collisions, and output mismatches reported before averages?
- Which claim label is supported:
formal-evidence,diagnostic-demo,construction-invalid, orclaim-overreach?