# 2. AI Mentor Mode Settings — Test Cases

User Type: **Super Administrator**
Source: *Mi Digital Academy - Education CRM Features Document*
Spec: ai_mentor_mode_settings.md — every feature, sub-feature, and rule covered

## Test Execution Policy

- Zero tolerance: any deviation from the documented behavior is a defect.
- Every failed test is logged with a Bug ID, the feature, the sub-feature, the expected vs actual result, and the severity; 100% of bugs are fixed before the group passes.
- 100% pass rate is required for the group to be marked complete.

## Coverage Matrix

| Feature | Sub-feature / Rule | Test IDs |
|---------|--------------------|----------|
| 2.1 Mentorship Style Configuration | Communication tone | TC-SA-AI-8-001 |
| 2.1 Mentorship Style Configuration | Check-in frequency (daily, weekly, per milestone) | TC-SA-AI-8-002 |
| 2.1 Mentorship Style Configuration | Goal-setting approach and feedback style | TC-SA-AI-8-003 |
| 2.1 Mentorship Style Configuration | Mentor persona per grade band | TC-SA-AI-8-004 |
| 2.1 Mentorship Style Configuration | Rule: no persona → default persona fallback | TC-SA-AI-8-005 |
| 2.1 Mentorship Style Configuration | Rule: check-in frequency is the maximum cadence | TC-SA-AI-8-006 |
| 2.1 Mentorship Style Configuration | Audit logging of mentorship style configuration | TC-SA-AI-8-007 |
| 2.2 Goal Tracking Settings | Goal types (exam, skill, habit) | TC-SA-AI-9-001 |
| 2.2 Goal Tracking Settings | Goal milestone definition (auto, manual) | TC-SA-AI-9-002 |
| 2.2 Goal Tracking Settings | Progress measurement per goal type | TC-SA-AI-9-003 |
| 2.2 Goal Tracking Settings | Goal review cadence and adjustment rules | TC-SA-AI-9-004 |
| 2.2 Goal Tracking Settings | Rule: no measurable metric → not trackable | TC-SA-AI-9-005 |
| 2.2 Goal Tracking Settings | Rule: re-target proposed and confirmed, not silent | TC-SA-AI-9-006 |
| 2.2 Goal Tracking Settings | Audit logging of goal tracking settings | TC-SA-AI-9-007 |
| 2.3 Study Habit Analysis Parameters | Habits tracked (time, consistency, balance, peak hours) | TC-SA-AI-10-001 |
| 2.3 Study Habit Analysis Parameters | Analysis window (7-day, 30-day rolling) | TC-SA-AI-10-002 |
| 2.3 Study Habit Analysis Parameters | Pattern detection (consistency, procrastination, cramming) | TC-SA-AI-10-003 |
| 2.3 Study Habit Analysis Parameters | Habit recommendations and report cadence | TC-SA-AI-10-004 |
| 2.3 Study Habit Analysis Parameters | Rule: insufficient activity → not analyzed | TC-SA-AI-10-005 |
| 2.3 Study Habit Analysis Parameters | Rule: one recommendation per pattern per cadence | TC-SA-AI-10-006 |
| 2.3 Study Habit Analysis Parameters | Audit logging of study habit analysis parameters | TC-SA-AI-10-007 |
| 2.4 Emotional Intelligence Thresholds | Sentiment detection sensitivity | TC-SA-AI-11-001 |
| 2.4 Emotional Intelligence Thresholds | Frustration indicator (errors, language, abandonment) | TC-SA-AI-11-002 |
| 2.4 Emotional Intelligence Thresholds | Motivation level assessment | TC-SA-AI-11-003 |
| 2.4 Emotional Intelligence Thresholds | Threshold per indicator and per grade band | TC-SA-AI-11-004 |
| 2.4 Emotional Intelligence Thresholds | Rule: one intervention per indicator per session | TC-SA-AI-11-005 |
| 2.4 Emotional Intelligence Thresholds | Rule: text input only, not silence | TC-SA-AI-11-006 |
| 2.4 Emotional Intelligence Thresholds | Audit logging of emotional intelligence thresholds | TC-SA-AI-11-007 |
| 2.5 Human Support Trigger Rules | Trigger conditions (low motivation, frustration, request) | TC-SA-AI-12-001 |
| 2.5 Human Support Trigger Rules | Support routing (counselor, parent, advisor) | TC-SA-AI-12-002 |
| 2.5 Human Support Trigger Rules | Escalation context (summary, indicators, history) | TC-SA-AI-12-003 |
| 2.5 Human Support Trigger Rules | Trigger cooldown and availability fallback | TC-SA-AI-12-004 |
| 2.5 Human Support Trigger Rules | Rule: student request bypasses cooldown | TC-SA-AI-12-005 |
| 2.5 Human Support Trigger Rules | Rule: suppressed trigger logged as suppressed | TC-SA-AI-12-006 |
| 2.5 Human Support Trigger Rules | Audit logging of human support trigger rules | TC-SA-AI-12-007 |

## 2.1 Mentorship Style Configuration

### TC-SA-AI-8-001 — Communication tone
**Type:** Positive
**Covers:** 2.1 → Communication tone
**Preconditions:** A Super Admin account is active; AI mentor mode is enabled.
**Steps:**
1. As a Super Admin, open AI Feature Configuration → AI Mentor Mode → Mentorship Style and set the communication tone (encouraging, structured, accountability-focused).
2. Verify the mentor's responses use the configured tone.
**Expected Result:** Communication tone — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-8-002 — Check-in frequency (daily, weekly, per milestone)
**Type:** Positive
**Covers:** 2.1 → Check-in frequency
**Preconditions:** Mentorship style is configured.
**Steps:**
1. Set the check-in frequency to daily and verify daily check-ins occur.
2. Set it to weekly and per milestone and verify the cadence matches.
**Expected Result:** Check-in frequency — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-8-003 — Goal-setting approach and feedback style
**Type:** Positive
**Covers:** 2.1 → Goal-setting and feedback
**Preconditions:** Mentorship style is configured.
**Steps:**
1. Set the goal-setting approach (SMART goals, milestone-based) and verify the mentor uses it.
2. Set the feedback style (immediate, batched, reflective) and verify feedback is delivered per the style.
**Expected Result:** Goal-setting approach and feedback style — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-8-004 — Mentor persona per grade band
**Type:** Positive
**Covers:** 2.1 → Mentor persona
**Preconditions:** Mentor personas are available.
**Steps:**
1. Select the mentor persona per grade band.
2. Verify a student in the band interacts with the assigned persona.
**Expected Result:** Mentor persona per grade band — delivered exactly as documented.
**Priority:** Medium

### TC-SA-AI-8-005 — Rule: no persona → default persona fallback
**Type:** Edge
**Covers:** 2.1 → Rule: persona fallback
**Preconditions:** A grade band has no assigned persona.
**Steps:**
1. Start a mentor session for a student in the band.
2. Verify the default persona is used.
**Expected Result:** A grade band with no assigned persona falls back to the default persona — delivered exactly as documented.
**Priority:** Medium

### TC-SA-AI-8-006 — Rule: check-in frequency is the maximum cadence
**Type:** Negative
**Covers:** 2.1 → Rule: cadence cap
**Preconditions:** The check-in frequency is set to weekly.
**Steps:**
1. Observe the mentor's check-ins over a week.
2. Verify the mentor does not check in more often than weekly.
**Expected Result:** The mentor does not check in more often than the configured frequency — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-8-007 — Audit logging of mentorship style configuration
**Type:** Positive
**Covers:** 2.1 → Audit logging
**Preconditions:** Mentorship style changes have been made.
**Steps:**
1. Open the audit trail and filter by mentorship style configuration.
2. Verify entries show the setting, the change, and the timestamp.
**Expected Result:** Audit logging of mentorship style configuration — delivered exactly as documented.
**Priority:** Critical

## 2.2 Goal Tracking Settings

### TC-SA-AI-9-001 — Goal types (exam, skill, habit)
**Type:** Positive
**Covers:** 2.2 → Goal types
**Preconditions:** A Super Admin is configuring goal tracking.
**Steps:**
1. Set the goal types (exam target, skill mastery, habit goal).
2. Verify each goal type is available for mentor-tracked goals.
**Expected Result:** Goal types — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-9-002 — Goal milestone definition (auto, manual)
**Type:** Positive
**Covers:** 2.2 → Milestone definition
**Preconditions:** Goal types are configured.
**Steps:**
1. Set the milestone definition to auto-generated and verify milestones are created for a new goal.
2. Set it to manual and verify milestones are set by the student/mentor.
**Expected Result:** Goal milestone definition — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-9-003 — Progress measurement per goal type
**Type:** Positive
**Covers:** 2.2 → Progress measurement
**Preconditions:** Goals exist with configured metrics.
**Steps:**
1. Set the progress measurement metric per goal type.
2. Verify goal progress is measured from the configured metric.
**Expected Result:** Progress measurement per goal type — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-9-004 — Goal review cadence and adjustment rules
**Type:** Positive
**Covers:** 2.2 → Review cadence and adjustments
**Preconditions:** Goals are being tracked.
**Steps:**
1. Define the goal review cadence (weekly, bi-weekly, monthly).
2. Set the adjustment rules (re-target on sustained under/over-performance) and verify they apply at review.
**Expected Result:** Goal review cadence and adjustment rules — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-9-005 — Rule: no measurable metric → not trackable
**Type:** Negative
**Covers:** 2.2 → Rule: measurability
**Preconditions:** A goal has no measurable metric.
**Steps:**
1. Attempt to track the goal.
2. Verify the mentor prompts for a measurable target (the goal is not tracked without one).
**Expected Result:** A goal with no measurable metric is not trackable — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-9-006 — Rule: re-target proposed and confirmed, not silent
**Type:** Negative
**Covers:** 2.2 → Rule: re-target confirmation
**Preconditions:** A goal is a re-target candidate (sustained under-performance).
**Steps:**
1. Trigger a re-target proposal.
2. Verify the re-target is proposed and not applied until the student confirms.
**Expected Result:** A goal re-target is proposed and confirmed by the student; it is not applied silently — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-9-007 — Audit logging of goal tracking settings
**Type:** Positive
**Covers:** 2.2 → Audit logging
**Preconditions:** Goal tracking setting changes have been made.
**Steps:**
1. Open the audit trail and filter by goal tracking settings.
2. Verify entries show the setting, the change, and the timestamp.
**Expected Result:** Audit logging of goal tracking settings — delivered exactly as documented.
**Priority:** Critical

## 2.3 Study Habit Analysis Parameters

### TC-SA-AI-10-001 — Habits tracked (time, consistency, balance, peak hours)
**Type:** Positive
**Covers:** 2.3 → Habits tracked
**Preconditions:** A Super Admin is configuring study habit analysis.
**Steps:**
1. Set the habits tracked (study time, consistency, subject balance, peak hours).
2. Verify each habit is analyzed for mentor-tracked students.
**Expected Result:** Habits tracked — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-10-002 — Analysis window (7-day, 30-day rolling)
**Type:** Positive
**Covers:** 2.3 → Analysis window
**Preconditions:** Study habit analysis is configured.
**Steps:**
1. Set the analysis window to 7-day rolling and verify the analysis uses the last 7 days.
2. Set it to 30-day rolling and verify the analysis uses the last 30 days.
**Expected Result:** Analysis window — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-10-003 — Pattern detection (consistency, procrastination, cramming)
**Type:** Positive
**Covers:** 2.3 → Pattern detection
**Preconditions:** Study habit analysis is configured; student activity exists.
**Steps:**
1. Define the pattern detection (consistency, procrastination, cramming).
2. Verify a student exhibiting a pattern has it detected.
**Expected Result:** Pattern detection — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-10-004 — Habit recommendations and report cadence
**Type:** Positive
**Covers:** 2.3 → Recommendations and cadence
**Preconditions:** Patterns are being detected.
**Steps:**
1. Set the habit recommendation rules (improvement suggestions per pattern).
2. Set the report cadence (weekly, monthly) and verify reports are delivered per the cadence.
**Expected Result:** Habit recommendations and report cadence — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-10-005 — Rule: insufficient activity → not analyzed
**Type:** Edge
**Covers:** 2.3 → Rule: sparse data
**Preconditions:** A student has insufficient activity in the analysis window.
**Steps:**
1. Run the study habit analysis for the student.
2. Verify the student is not analyzed (no false patterns from sparse data).
**Expected Result:** A student with insufficient activity in the window is not analyzed — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-10-006 — Rule: one recommendation per pattern per cadence
**Type:** Negative
**Covers:** 2.3 → Rule: no repetition
**Preconditions:** A pattern is detected for a student.
**Steps:**
1. Deliver the habit report for the cadence period.
2. Verify the pattern produces at most one recommendation in the report (no repetition).
**Expected Result:** A detected pattern produces at most one recommendation per pattern per report cadence — delivered exactly as documented.
**Priority:** Medium

### TC-SA-AI-10-007 — Audit logging of study habit analysis parameters
**Type:** Positive
**Covers:** 2.3 → Audit logging
**Preconditions:** Study habit analysis parameter changes have been made.
**Steps:**
1. Open the audit trail and filter by study habit analysis parameters.
2. Verify entries show the parameter, the change, and the timestamp.
**Expected Result:** Audit logging of study habit analysis parameters — delivered exactly as documented.
**Priority:** Critical

## 2.4 Emotional Intelligence Thresholds

### TC-SA-AI-11-001 — Sentiment detection sensitivity
**Type:** Positive
**Covers:** 2.4 → Sentiment sensitivity
**Preconditions:** A Super Admin is configuring emotional intelligence.
**Steps:**
1. Set the sentiment detection sensitivity (low, medium, high).
2. Verify sentiment is detected per the configured sensitivity.
**Expected Result:** Sentiment detection sensitivity — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-11-002 — Frustration indicator (errors, language, abandonment)
**Type:** Positive
**Covers:** 2.4 → Frustration indicator
**Preconditions:** Emotional intelligence is configured.
**Steps:**
1. Configure the frustration indicator (repeated errors, negative language, session abandonment).
2. Trigger each indicator and verify frustration is detected.
**Expected Result:** Frustration indicator — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-11-003 — Motivation level assessment
**Type:** Positive
**Covers:** 2.4 → Motivation assessment
**Preconditions:** Emotional intelligence is configured.
**Steps:**
1. Set the motivation level assessment (engagement signals, goal progress).
2. Verify the motivation level is assessed from the configured signals.
**Expected Result:** Motivation level assessment — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-11-004 — Threshold per indicator and per grade band
**Type:** Positive
**Covers:** 2.4 → Thresholds
**Preconditions:** Emotional intelligence is configured.
**Steps:**
1. Define the threshold per indicator (trigger level).
2. Set a per-grade-band threshold (age-appropriate sensitivity) and verify it applies to the band.
**Expected Result:** Threshold per indicator and per grade band — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-11-005 — Rule: one intervention per indicator per session
**Type:** Negative
**Covers:** 2.4 → Rule: intervention cap
**Preconditions:** An indicator threshold is triggered repeatedly in a session.
**Steps:**
1. Trigger the same indicator multiple times in one session.
2. Verify at most one intervention occurs per indicator per session.
**Expected Result:** A threshold trigger produces at most one intervention per indicator per session — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-11-006 — Rule: text input only, not silence
**Type:** Negative
**Covers:** 2.4 → Rule: input scope
**Preconditions:** A student is silent (no text input) in a session.
**Steps:**
1. Observe the sentiment detection during the silence.
2. Verify no sentiment is inferred from silence alone.
**Expected Result:** The sentiment detection operates on the student's text input only; it does not infer from silence alone — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-11-007 — Audit logging of emotional intelligence thresholds
**Type:** Positive
**Covers:** 2.4 → Audit logging
**Preconditions:** Emotional intelligence threshold changes have been made.
**Steps:**
1. Open the audit trail and filter by emotional intelligence thresholds.
2. Verify entries show the threshold, the change, and the timestamp.
**Expected Result:** Audit logging of emotional intelligence thresholds — delivered exactly as documented.
**Priority:** Critical

## 2.5 Human Support Trigger Rules

### TC-SA-AI-12-001 — Trigger conditions (low motivation, frustration, request)
**Type:** Positive
**Covers:** 2.5 → Trigger conditions
**Preconditions:** A Super Admin is configuring human support triggers.
**Steps:**
1. Set the trigger conditions (sustained low motivation, repeated frustration, student request).
2. Trigger each condition and verify a human support trigger fires.
**Expected Result:** Trigger conditions — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-12-002 — Support routing (counselor, parent, advisor)
**Type:** Positive
**Covers:** 2.5 → Support routing
**Preconditions:** Trigger conditions are configured; support recipients exist.
**Steps:**
1. Configure the support routing (counselor, parent notification, academic advisor).
2. Trigger each condition and verify it routes to the configured recipient.
**Expected Result:** Support routing — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-12-003 — Escalation context (summary, indicators, history)
**Type:** Positive
**Covers:** 2.5 → Escalation context
**Preconditions:** A human support trigger fires.
**Steps:**
1. Trigger a human support escalation.
2. Verify the escalation context (session summary, indicator values, student history) is generated and visible to the recipient.
**Expected Result:** Escalation context — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-12-004 — Trigger cooldown and availability fallback
**Type:** Positive
**Covers:** 2.5 → Cooldown and fallback
**Preconditions:** Trigger rules are configured.
**Steps:**
1. Set the trigger cooldown (minimum interval between triggers per student).
2. Set the availability window with fallback and verify a trigger outside the window falls back.
**Expected Result:** Trigger cooldown and availability fallback — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-12-005 — Rule: student request bypasses cooldown
**Type:** Negative
**Covers:** 2.5 → Rule: request bypass
**Preconditions:** A student is within the trigger cooldown window.
**Steps:**
1. Request support as the student.
2. Verify the trigger succeeds (not blocked by the cooldown).
**Expected Result:** A student-requested support trigger always succeeds; it is never blocked by the cooldown — delivered exactly as documented.
**Priority:** Critical

### TC-SA-AI-12-006 — Rule: suppressed trigger logged as suppressed
**Type:** Negative
**Covers:** 2.5 → Rule: suppression logging
**Preconditions:** An indicator-based trigger fires within the cooldown window.
**Steps:**
1. Trigger an indicator-based escalation within the cooldown.
2. Verify the trigger is suppressed and logged as suppressed.
**Expected Result:** A trigger within the cooldown window is suppressed for indicator-based triggers but logged as suppressed — delivered exactly as documented.
**Priority:** High

### TC-SA-AI-12-007 — Audit logging of human support trigger rules
**Type:** Positive
**Covers:** 2.5 → Audit logging
**Preconditions:** Human support trigger rule changes have been made.
**Steps:**
1. Open the audit trail and filter by human support trigger rules.
2. Verify entries show the rule, the change, and the timestamp.
**Expected Result:** Audit logging of human support trigger rules — delivered exactly as documented.
**Priority:** Critical