{
  "context": "Tests whether the agent correctly applies the classification system across three packets with different failure modes: Alpha (all checks pass = CLEAN), Beta (stale timestamp = OPERATIONAL), Gamma (empty required fields = OPERATIONAL), producing structured output with per-check summaries and recovery steps for each.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Alpha classified CLEAN",
      "description": "Handoff-alpha is classified as CLEAN (all checks pass)",
      "max_score": 12
    },
    {
      "name": "Beta classified OPERATIONAL",
      "description": "Handoff-beta is classified as OPERATIONAL (not CLEAN, not CRITICAL) due to the stale timestamp",
      "max_score": 12
    },
    {
      "name": "Gamma classified OPERATIONAL",
      "description": "Handoff-gamma is classified as OPERATIONAL (not CLEAN, not CRITICAL) due to empty required fields",
      "max_score": 12
    },
    {
      "name": "Per-check breakdown for all three",
      "description": "Each handoff packet has its own per-check summary showing pass/fail for each check performed",
      "max_score": 14
    },
    {
      "name": "Recovery steps for Beta and Gamma",
      "description": "Both handoff-beta and handoff-gamma have non-empty recovery steps; handoff-alpha states no recovery steps are required",
      "max_score": 12
    },
    {
      "name": "Escalation for all three",
      "description": "All three packets include an escalation recommendation section (even if Alpha's recommendation is 'no escalation needed')",
      "max_score": 10
    },
    {
      "name": "CLEAN/OPERATIONAL/CRITICAL labels used",
      "description": "All three classifications use explicitly one of the labels CLEAN, OPERATIONAL, or CRITICAL",
      "max_score": 14
    },
    {
      "name": "Freshness check on all packets",
      "description": "The freshness of updated_at is checked and reported for all three packets",
      "max_score": 12
    },
    {
      "name": "Summary table or section",
      "description": "Report concludes with a summary that shows the classification of all three packets in one place",
      "max_score": 2
    }
  ]
}