> ## Documentation Index
> Fetch the complete documentation index at: https://docs.crazygoldfish.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Evaluation Data → Action Plans

> See how exam and assignment evaluation results flow into Student, Teacher, and Parent Action Plan APIs to close the loop between assessment and remediation.

<Info>
  The **Evaluation Layer** produces scores, feedback, and annotated responses.\
  The **Personalization Layer** transforms these into **Student, Teacher, and Parent Action Plans**, ensuring assessment drives measurable improvement.
</Info>

***

## What Evaluation Produces

* **Scores & Totals** → per student, per question.
* **Feedback JSON** → strengths, improvements, rubric alignment.
* **Annotated Copies** → transparent student/parent view.
* **Dashboards** → turnaround time, accuracy, compliance.

***

## How Data Flows Into Action Plans

<Steps>
  <Step title="Student Action Plans">
    * Use scores + misconceptions to generate **personalised targets, tasks, and milestones**.
    * Plans feed into **AI Studio** for remedial worksheets and practice packs.
  </Step>

  <Step title="Teacher Action Plans">
    * Aggregate results to identify **common errors** and group students.
    * Recommend **reteach strategies** + extension content for advanced learners.
  </Step>

  <Step title="Parent Action Plans">
    * Simplify scores and feedback into **weekly activity packs**.
    * Provide clear, non-technical progress reports for at-home support.
  </Step>
</Steps>

***

## Stakeholder Impact

* **Students** → Receive clear learning pathways instead of just grades.
* **Teachers** → Get grouped insights and reteach strategies.
* **Parents** → Stay informed with simple activity packs.
* **Leaders** → See remediation trends alongside evaluation dashboards.

***

## Ecosystem Integration

* **Upstream:** Exam & Assignment Evaluation Suites.
* **Core:** Action Plan APIs (Student, Teacher, Parent).
* **Downstream:** AI Studio generates targeted content → ClassTrack refines in classrooms.

***

## Next Step

**Next →** See how [Action Plans → Content Generation](/products/integration-eco-system/action-plans) to explore how plans automatically generate lesson plans and worksheets in AI Studio.

## FAQ

<script
  type="application/ld+json"
  dangerouslySetInnerHTML={{
__html: JSON.stringify(
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
    {
        "@type": "Question",
        "name": "How do we plug the Evaluation capability into our LMS/ERP—should we use the embeddable UI or the APIs, and how is auth handled?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "You can integrate via a white‑label embeddable UI (iframe/link) for a fast, branded experience, or use the Evaluation APIs for full programmatic control. Authentication is token‑based, with role‑based tokens to separate sandbox and production access. Both surfaces produce the same structured outputs so you can sync results into existing LMS/ERP workflows."
        }
    },
    {
        "@type": "Question",
        "name": "What does the Final Results API return for each submission in the Evaluation flow?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "The Final Results API returns exam‑level scores, section summaries, model answers, step‑wise marking, and teacher feedback, including human review fields like isApproved. You can also retrieve feedback JSON and annotated copies, enabling rich reports and auditable artefacts for multimodal work."
        }
    },
    {
        "@type": "Question",
        "name": "How do we submit multimodal inputs like handwriting scans, diagrams, audio, or video for evaluation?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "Submit PDFs/images for handwriting and diagrams, or audio/video files alongside a model answer and/or rubric. The system normalizes multimodal inputs and applies rubric‑based and model‑answer grading to deliver consistent scoring and feedback. Outputs are structured for easy consumption downstream."
        }
    },
    {
        "@type": "Question",
        "name": "How are human‑in‑the‑loop reviews, re‑checks, and overrides represented in the integration?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "Teachers remain a human‑in‑the‑loop, with workflows for re‑checks and overrides before publishing results. Review states and fields such as isApproved are exposed so you can track and enforce final approval. Audit logs capture revisions for transparency and governance."
        }
    },
    {
        "@type": "Question",
        "name": "Can we receive asynchronous notifications when an evaluation completes?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "Yes. Subscribe to webhooks to receive completion events and then fetch the full payload via the Final Results API. This pattern decouples long‑running evaluations from your UI while preserving secure access via role‑based tokens."
        }
    },
    {
        "@type": "Question",
        "name": "How does the Evaluation stack ensure CBSE/ICSE curriculum alignment and GDPR‑aligned data handling?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "Evaluation and reporting are CBSE/ICSE/GDPR aligned, with configurable rubrics and board mappings. Transparent breakdowns (e.g., section scores and step‑wise marking) and audit‑friendly logs support institutional compliance. Role‑based access and export controls help keep data governance intact."
        }
    },
    {
        "@type": "Question",
        "name": "What deliverables can we sync to our dashboards or data lake from the Evaluation workflow?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "You can ingest feedback JSON, scores, section summaries, and annotated copies into your analytics or LMS/ERP. Exportable PDFs/CSVs and dashboard‑ready summaries make it simple to surface results to teachers, leaders, or parents while preserving auditability."
        }
    },
    {
        "@type": "Question",
        "name": "How do rubric‑based and model‑answer grading modes work together in one integration?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "The platform supports dual modes—rubric‑based and model‑answer grading—and you can use them independently or in combination per item. Results include step‑wise marking and objective error signals where applicable, providing consistent, explainable scoring across subjective and objective tasks."
        }
    },
    {
        "@type": "Question",
        "name": "What does an end‑to‑end Evaluation integration look like from submission to downstream action?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "Your app submits multimodal responses with rubrics/model answers, triggers evaluation, and listens on webhooks for completion. It then retrieves scores, section summaries, step‑wise marking, feedback JSON, and annotated copies via the Final Results API. These insights can feed Personalize/Generate modules to produce action plans, lesson plans, and worksheets."
        }
    },
    {
        "@type": "Question",
        "name": "How are rechecks and audit trails exposed for governance and accuracy monitoring?",
        "acceptedAnswer": {
            "@type": "Answer",
            "text": "Auditable workflows support re‑checks and teacher overrides, with logs that record review actions and publication states. Live dashboards and exports provide visibility into turnaround and accuracy, while structured payloads ensure consistent, reviewable evidence at scale."
        }
    }
]
}
)
}}
/>

<AccordionGroup>
  <Accordion title="How do we plug the Evaluation capability into our LMS/ERP—should we use the embeddable UI or the APIs, and how is auth handled?">
    You can integrate via a white‑label embeddable UI (iframe/link) for a fast, branded experience, or use the Evaluation APIs for full programmatic control. Authentication is token‑based, with role‑based tokens to separate sandbox and production access. Both surfaces produce the same structured outputs so you can sync results into existing LMS/ERP workflows.
  </Accordion>

  <Accordion title="What does the Final Results API return for each submission in the Evaluation flow?">
    The Final Results API returns exam‑level scores, section summaries, model answers, step‑wise marking, and teacher feedback, including human review fields like isApproved. You can also retrieve feedback JSON and annotated copies, enabling rich reports and auditable artefacts for multimodal work.
  </Accordion>

  <Accordion title="How do we submit multimodal inputs like handwriting scans, diagrams, audio, or video for evaluation?">
    Submit PDFs/images for handwriting and diagrams, or audio/video files alongside a model answer and/or rubric. The system normalizes multimodal inputs and applies rubric‑based and model‑answer grading to deliver consistent scoring and feedback. Outputs are structured for easy consumption downstream.
  </Accordion>

  <Accordion title="How are human‑in‑the‑loop reviews, re‑checks, and overrides represented in the integration?">
    Teachers remain a human‑in‑the‑loop, with workflows for re‑checks and overrides before publishing results. Review states and fields such as isApproved are exposed so you can track and enforce final approval. Audit logs capture revisions for transparency and governance.
  </Accordion>

  <Accordion title="Can we receive asynchronous notifications when an evaluation completes?">
    Yes. Subscribe to webhooks to receive completion events and then fetch the full payload via the Final Results API. This pattern decouples long‑running evaluations from your UI while preserving secure access via role‑based tokens.
  </Accordion>

  <Accordion title="How does the Evaluation stack ensure CBSE/ICSE curriculum alignment and GDPR‑aligned data handling?">
    Evaluation and reporting are CBSE/ICSE/GDPR aligned, with configurable rubrics and board mappings. Transparent breakdowns (e.g., section scores and step‑wise marking) and audit‑friendly logs support institutional compliance. Role‑based access and export controls help keep data governance intact.
  </Accordion>

  <Accordion title="What deliverables can we sync to our dashboards or data lake from the Evaluation workflow?">
    You can ingest feedback JSON, scores, section summaries, and annotated copies into your analytics or LMS/ERP. Exportable PDFs/CSVs and dashboard‑ready summaries make it simple to surface results to teachers, leaders, or parents while preserving auditability.
  </Accordion>

  <Accordion title="How do rubric‑based and model‑answer grading modes work together in one integration?">
    The platform supports dual modes—rubric‑based and model‑answer grading—and you can use them independently or in combination per item. Results include step‑wise marking and objective error signals where applicable, providing consistent, explainable scoring across subjective and objective tasks.
  </Accordion>

  <Accordion title="What does an end‑to‑end Evaluation integration look like from submission to downstream action?">
    Your app submits multimodal responses with rubrics/model answers, triggers evaluation, and listens on webhooks for completion. It then retrieves scores, section summaries, step‑wise marking, feedback JSON, and annotated copies via the Final Results API. These insights can feed Personalize/Generate modules to produce action plans, lesson plans, and worksheets.
  </Accordion>

  <Accordion title="How are rechecks and audit trails exposed for governance and accuracy monitoring?">
    Auditable workflows support re‑checks and teacher overrides, with logs that record review actions and publication states. Live dashboards and exports provide visibility into turnaround and accuracy, while structured payloads ensure consistent, reviewable evidence at scale.
  </Accordion>
</AccordionGroup>