Cohort progression signal and student-success support plan

Elite LLM Handoff

Explain the support_cohort_progression recommendation using the governed Fortress output and caveats.

governed_handoff:
  - Goal: support_cohort_progression
  - Primary path: cohort_support_plan_started, score 0.7296, high confidence
  - Supporting signals: advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster
  - Runtime relationship sets returned: 8
  - Domain journeys: 120; events: 480; relationship sets: 120

Outcome Prompt JSON

{
  "schema": "kynticai.elite.outcome-prompts.v1",
  "caseStudyId": "education-student-success-cohort",
  "domain": "Education / university operations",
  "stakeholder": "Student success lead",
  "runtimeQuery": "A student cohort shows attendance drop, LMS engagement decline, missed assessment checkpoints, and clustered support tickets. What student-success operations action should happen next?",
  "promptContract": "You are Elite, the KynticAI recommendation-writing layer.\n\nYou receive governed Fortress output, not raw customer data. Use only the supplied relationship-set facts, scores, confidence labels, alternatives, caveats, and outcome goal.\n\nWrite a decisive operational recommendation for the accountable team. Your output must include:\n1. Recommended action.\n2. Why this action is supported by the comparable relationship sets.\n3. What to do now.\n4. When to choose one of the alternative actions.\n5. What outcome to capture so the system improves.\n\nRules:\n- Do not invent facts outside the governed handoff.\n- Do not promise revenue, retention, clinical, legal, financial, safety, or compliance outcomes.\n- Do not apologise for using synthetic demo data.\n- Be commercially useful, specific, and direct.\n- Keep human review and claim boundaries visible where the domain requires them.",
  "governedFacts": [
    "Goal: support_cohort_progression",
    "Primary path: cohort_support_plan_started, score 0.7296, high confidence",
    "Supporting signals: advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster",
    "Runtime relationship sets returned: 8",
    "Domain journeys: 120; events: 480; relationship sets: 120"
  ],
  "outcomePrompts": [
    {
      "rank": 1,
      "outcome": "cohort_support_plan_started",
      "action": "Start a cohort support plan with advisor outreach, assessment-deadline triage, and targeted workshop invitations.",
      "confidence": "High",
      "score": "0.7296",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Cohort progression signal and student-success support plan\nDomain: Education / university operations\nStakeholder: Student success lead\nRuntime query: A student cohort shows attendance drop, LMS engagement decline, missed assessment checkpoints, and clustered support tickets. What student-success operations action should happen next?\n\nGoverned facts:\n- Goal: support_cohort_progression\n- Primary path: cohort_support_plan_started, score 0.7296, high confidence\n- Supporting signals: advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 480; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 1\n- Outcome: cohort_support_plan_started\n- Action: Start a cohort support plan with advisor outreach, assessment-deadline triage, and targeted workshop invitations.\n- Confidence: High\n- Score: 0.7296\n- Evidence summary: Comparable relationship sets ended in cohort_support_plan_started and share signals such as advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster.\n- Supporting relationship sets: set-6586ee813dc3, set-dab1c06dee9a, set-73036cb50e8a, set-5f63cb4e4d2f, set-442385d44256\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Recommended action: Start a cohort support plan with advisor outreach, assessment-deadline triage, and targeted workshop invitations.\n\nWhy this is supported: Fortress ranked cohort_support_plan_started first with score 0.7296 and high confidence. Comparable relationship sets ended in cohort_support_plan_started and share signals such as advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster. The supporting relationship sets are set-6586ee813dc3, set-dab1c06dee9a, set-73036cb50e8a, set-5f63cb4e4d2f, set-442385d44256.\n\nWhat to do now: brief the student success lead with the runtime query, the shared signals (advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster), and the exact action. Keep the action specific to this scenario rather than turning it into a generic campaign or workflow rule.\n\nWhen to choose an alternative: switch if the live evidence is closer to workshop_registered (Invite the cohort to the targeted assessment workshop and capture registration before escalating individual interventions.); advisor_contact_completed (Prioritise advisor contact for students with repeated engagement gaps before sending broad course communications.).\n\nOutcome to capture: record the action taken, the reviewed outcome, and any contradictory signal so the next Scout/Fortress comparison has stronger evidence."
    },
    {
      "rank": 2,
      "outcome": "workshop_registered",
      "action": "Invite the cohort to the targeted assessment workshop and capture registration before escalating individual interventions.",
      "confidence": "Medium",
      "score": "0.3614",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Cohort progression signal and student-success support plan\nDomain: Education / university operations\nStakeholder: Student success lead\nRuntime query: A student cohort shows attendance drop, LMS engagement decline, missed assessment checkpoints, and clustered support tickets. What student-success operations action should happen next?\n\nGoverned facts:\n- Goal: support_cohort_progression\n- Primary path: cohort_support_plan_started, score 0.7296, high confidence\n- Supporting signals: advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 480; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 2\n- Outcome: workshop_registered\n- Action: Invite the cohort to the targeted assessment workshop and capture registration before escalating individual interventions.\n- Confidence: Medium\n- Score: 0.3614\n- Evidence summary: Comparable relationship sets ended in workshop_registered and share signals such as assessment_workshop, registration_completed, targeted_invite.\n- Supporting relationship sets: set-317a95076056\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Invite the cohort to the targeted assessment workshop and capture registration before escalating individual interventions.\n\nWhy this route exists: Fortress returned workshop_registered at rank 2 with score 0.3614 and medium confidence. Comparable relationship sets ended in workshop_registered and share signals such as assessment_workshop, registration_completed, targeted_invite. The supporting relationship sets are set-317a95076056.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches assessment_workshop, registration_completed, targeted_invite more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the student success lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether workshop_registered happened and feed that reviewed result back into the relationship-set evidence."
    },
    {
      "rank": 3,
      "outcome": "advisor_contact_completed",
      "action": "Prioritise advisor contact for students with repeated engagement gaps before sending broad course communications.",
      "confidence": "Low",
      "score": "0.1949",
      "prompt": "You are Elite, the KynticAI recommendation-writing layer.\n\nUse only the governed Fortress handoff below. Do not invent facts.\n\nCase study: Cohort progression signal and student-success support plan\nDomain: Education / university operations\nStakeholder: Student success lead\nRuntime query: A student cohort shows attendance drop, LMS engagement decline, missed assessment checkpoints, and clustered support tickets. What student-success operations action should happen next?\n\nGoverned facts:\n- Goal: support_cohort_progression\n- Primary path: cohort_support_plan_started, score 0.7296, high confidence\n- Supporting signals: advisor_outreach, assessment_checkpoint_missed, attendance_drop, lms_engagement_decline, support_ticket_cluster\n- Runtime relationship sets returned: 8\n- Domain journeys: 120; events: 480; relationship sets: 120\n\nCandidate outcome to explain:\n- Rank: 3\n- Outcome: advisor_contact_completed\n- Action: Prioritise advisor contact for students with repeated engagement gaps before sending broad course communications.\n- Confidence: Low\n- Score: 0.1949\n- Evidence summary: Comparable relationship sets ended in advisor_contact_completed and share signals such as advisor_queue, contact_completed, repeated_engagement_gap.\n- Supporting relationship sets: set-498cb3a7a884\n\nWrite the operational recommendation for this outcome.\nInclude why the comparable relationship sets support it, what the team should do now, when this outcome should be chosen over the other ranked options, and what reviewed outcome must be captured next.",
      "expectedTextOutput": "Alternative action: Prioritise advisor contact for students with repeated engagement gaps before sending broad course communications.\n\nWhy this route exists: Fortress returned advisor_contact_completed at rank 3 with score 0.1949 and low confidence. Comparable relationship sets ended in advisor_contact_completed and share signals such as advisor_queue, contact_completed, repeated_engagement_gap. The supporting relationship sets are set-498cb3a7a884.\n\nWhen to use it: choose this instead of the primary route when the live evidence matches advisor_queue, contact_completed, repeated_engagement_gap more closely than the current top-ranked outcome.\n\nWhat to do now: make the action explicit for the student success lead, preserve the caveat for this domain, and avoid inventing extra facts beyond the governed handoff.\n\nOutcome to capture: record whether advisor_contact_completed happened and feed that reviewed result back into the relationship-set evidence."
    }
  ]
}