// ACTION.PLANS
One Evaluation. Three Action Plans. Every Student Covered.
From one evaluation event: a student remediation plan, a tiered teacher intervention, and a parent activity pack — structured, simultaneous, delivered via API.
// EVAL.COMPLETE → PLANS.GENERATED
eval_id: EVAL_8821
students_eval: 32
questions_eval: 26
// FIELD.VALIDATION
// THE.SYSTEM
Personalisation Infrastructure, Not a Recommendation Layer
STUDENT.PLAN
action_types: Remedial · Reinforcement · Challenge
per_question: ✓
content_routing: video | worksheet | MCQ
confidence_booster: ✓ included
TEACHER.PLAN
tier_1: whole class reteach
tier_2: small group + student IDs
tier_3: individual remediation
coverage: 100% students mapped
PARENT.PLAN
format: plain language
gap_context: ✓ included
activity_pack: Content Builder API
trigger: auto on eval complete
// all three plans generated simultaneously — delivered via API at eval completion · part of evaluation infrastructure
// CAPABILITIES
Three Plans. Three Action Frameworks. One Pipeline.
STUDENT.PLAN
Individual Learning Direction
→ Remedial, Reinforcement, or Challenge per question
→ Overall insight + confidence booster tip
→ Content type tag per action (video / worksheet / MCQ)
TEACHER.PLAN
Tiered Class Intervention
→ Tier 1: whole-class reteach
→ Tier 2: small group with specific student IDs
→ Tier 3: individual remediation
PARENT.PLAN
Engagement Activity Pack
→ Plain-language gap explanation
→ At-home activity pack via Content Builder API
→ Aligned to student's actual evaluated gaps
Content Routing Built In
Every action outputs a content_type tag — video, worksheet, MCQ, activity pack — your content system can consume directly via API.
100% Student Coverage by Design
Every student in the evaluation is mapped to at least one teacher plan action. Coverage is a design constraint, not a best-effort outcome.
// HOW.IT.WORKS
From One Evaluation Signal to Three Structured Outputs
01
Signal Capture
Per-criterion evaluation — score, confidence, gap flags, error type
02
Classification
Remedial (delta ≤3) · Reinforcement (3–7) · Challenge (>7)
03
Simultaneous Plan Generation
Student plan, teacher plan, parent plan — generated in parallel
04
Structured Learning Intelligence Output
All three plans returned via API — routed per stakeholder destination
STUDENT.PLAN ✓
answer_sheet_id: c0f312dc...7ae7
overall_action_type: Remediation
Q5 → Remediation — review separation techniques
Q7 → Reinforcement — revisit mixture properties
confidence_booster: ✓ included
TEACHER.PLAN ✓
concept: Separation Techniques
tier_1: whole-class reteach (18 students)
tier_2: small group — [S04, S07, S12, S19]
tier_3: individual — [S07]
coverage: 32/32 students mapped
PARENT.PLAN ✓
student: Govardhani
gap_context: separation techniques, mixture properties
activity_pack: GENERATED
via: Content Builder API
// WHY.THIS.WORKS
Built for Platforms Where Learning Direction Decisions Matter.
HUMAN.IN.THE.LOOP
Teacher is the control point.
AI handles volume. Teachers govern outcomes. Every evaluation is human-in-the-loop by design — not an optional setting, a structural requirement built into the pipeline. The rubric is yours. The override is yours. The publish gate is yours.
FULL.AUDIT.TRAIL
Every evaluation is auditable.
Full signal trail on every response. Every override logged with reviewer ID, timestamp, and delta. Every decision traceable to the evaluation that triggered it. Nothing moves through the system without a record.
INSTITUTIONAL.VALIDATION
Validated at India AI Impact Summit 2026.
Published in the MeitY Compendium. Results independently reviewed and presented at national level — not self-reported benchmarks. Third-party institutional credibility with a public record.
HUMAN.IN.THE.LOOP
Policy-aware. Human-verified. Infrastructure-grade.
CGF's evaluation system is human-in-the-loop by design. Teachers remain the control point — AI handles volume, humans handle governance. This is not an automation layer. It is evaluation infrastructure with built-in human authority at every control point.
// GOVERNANCE
Human-in-the-Loop at Every Control Point
Action plans built on weak signals cause real harm. A teacher running the wrong intervention loses class time. A student practising a mastered concept isn't closing the gap they actually need to close. CGF's governance layer gates every low-confidence evaluation through human review before it enters the plan generation pipeline.
Confidence-Gated Plan Delivery
→ plans withheld below confidence threshold
→ human review triggered automatically
→ no plan delivered without signal lock
Teacher Override at Every Stage
→ action type editable per student
→ plan suppressed or escalated by teacher
→ override logged to full audit trail
Full Signal-to-Plan Audit
→ eval signal → plan output logged
→ every override recorded with timestamp
→ exportable per student / per cohort
// USE.CASES
Find Your Action Plans Use Case
EdTech Platforms
Student, teacher, and parent plans — routed to wherever each stakeholder lives on your platform.
signal-to-plan loop across millions of students, no manual intervention
See use case →
Coaching & Test-Prep Networks
Gap-targeted plans for every JEE, NEET, or UPSC response.
per-student action plans from every subjective evaluation
See use case →
LMS / ERP Platforms
Three structured plan outputs your systems can act on directly via API.
structured learning intelligence embedded directly into the teacher dashboard
See use case →
Institutions & Public Programs
Student, teacher, and parent action plans at program scale.
programme-wide action plans with full audit trail and parent communication
See use case →
// COMMON.QUESTIONS
CGF's Action Plans product generates three simultaneous plans from every evaluation — a Student Action Plan (Remedial, Reinforcement, Challenge), a Teacher Action Plan (tiered class intervention by concept), and a Parent Engagement Activity Pack. All three use the same evaluation signals and are delivered via API. Scoring reliability: 93% ICC.
What are the three Action Plans CGF generates?
CGF generates three plans simultaneously from every evaluation: a Student Action Plan organising gaps by action type (Remedial, Reinforcement, Challenge), a Teacher Action Plan with tiered class interventions grouped by concept and student ID, and a Parent Action Plan translating evaluation signals into plain-language context and a structured activity pack generated via CGF's Content Builder APIs. All three are delivered via API at evaluation completion.
What is the difference between Remedial, Reinforcement, and Challenge actions?
CGF classifies every evaluated gap by action type based on the confidence delta — the measured gap between expected and actual performance. Remedial actions target concepts where the student needs to rebuild foundational understanding. Reinforcement targets concepts with partial grasp needing consolidation. Challenge targets concepts where the student is performing strongly and is ready to go further. The plan is a complete picture of where the student is and where they can go next.
How does the teacher plan group students for intervention?
The teacher plan uses a three-tier structure. Tier 1 covers whole-class re-teaching for concepts most of the class is struggling with. Tier 2 groups students who share the same misconception for small-group instruction, specifying which student IDs belong to each group. Tier 3 assigns individual remediation to the students who need the most intensive support. Every student in the class is mapped to at least one action — by design.
What is the Parent Engagement Activity Pack?
The Parent Action Plan translates a student's evaluation signals into plain-language context — what the student is working on, what the specific gap looks like, and what a parent can do at home. It connects to CGF's Content Builder API, which generates a subject-specific Parent Engagement Activity Pack: structured activities aligned to the student's actual evaluated gaps. Not a generic homework reminder — a content output grounded in the evaluation signal.
How accurate are the signals that drive action plans?
CGF achieves 93% reliability in total scoring alignment (ICC), ensuring gap signals reflect actual learner performance before they influence any plan. Structured feedback signals are delivered with 90%+ accuracy instantly. Low-confidence evaluations are held for human review before entering the plan generation pipeline — ensuring plans are only generated from signals you can trust.
// NEXT.STEP
Close the Loop for Every Stakeholder, Every Evaluation
Student direction. Teacher intelligence. Parent engagement. Three structured action plans — generated simultaneously from evaluation signals, connected to content generation, at any scale, via API.