$ trace-eval run examples/hermes_good.jsonl

============================================================
  TRACE-EVAL SCORECARD  Total: 98.9/100
============================================================

DIMENSION SCORES:
  reliability           100.0  (high)
  efficiency             94.5  (medium)
  retrieval             100.0  (high)
  tool_discipline       100.0  (high)
  context               100.0  (high)

ADAPTER CAPABILITY REPORT:
  has_span_ids: False
  has_parent_spans: False
  has_event_latency: True
  has_token_data: True
  has_tool_calls: True
  has_cost_data: True
  has_retrieval_fields: True
