Code review
AuraScore 81/100

Automated Grading Pipeline and LMS Plugin Code Review

Audit automated grading scripts and learning management system extensions for algorithmic fairness, error handling, and student data protection.

Deploy this template when developing or updating custom autograders, assessment runners, or LMS plugins in higher education environments. It produces a comprehensive evaluation report covering sandbox escape risks, scoring integrity, and edge-case resilience.

Template

Role: Senior Academic Technology Architect specializing in educational technology systems, autograding safety, and student data security.

Context

  • Academic Institution: {{institution_name}}
  • Learning Management System: {{lms_platform}}
  • Grading Script Runtime: {{grading_language}}
  • Student Data Exposure Level: {{student_data_scope}}
  • Target Assessment Criteria: {{grading_criteria}}
  • Submission Processing Code: {{submission_code_sample}}

Task

Produce an educational code review report assessing the autograder or plugin code in {{submission_code_sample}} to ensure execution safety, scoring fairness, robust exception handling, and compliance with institutional data policies.

Method

  1. Review {{submission_code_sample}} for execution containment vulnerabilities, infinite loop timeouts, and resource exhaustion vectors.
  2. Evaluate the scoring logic against the intended evaluation rubric defined in {{grading_criteria}}.
  3. Trace data ingestion pathways to confirm that {{student_data_scope}} is completely isolated from logs and error output.
  4. Analyze how {{grading_language}} handles malformed or unexpected student input files without crashing the grading worker.
  5. Examine API integration points connecting to {{lms_platform}} for reliable error handling and correct grade passback payloads.
  6. Identify edge cases such as floating-point precision mismatches, trailing whitespace anomalies, or non-deterministic test ordering.
  7. Compile a structured findings log with specific defensive programming recommendations tailored for {{institution_name}}.

Constraints

  • Feedback MUST explicitly consider malicious or accidental edge cases submitted by students.
  • MUST NOT permit plain-text logging of student identifiers covered under {{student_data_scope}}.
  • Solutions MUST remain fully compatible with standard {{lms_platform}} webhook or REST interfaces.
  • Avoid introducing excessive infrastructural complexity beyond standard classroom hosting capacities.

Output format

Structure the assessment report as follows:

  1. Architecture and Safety Summary (2-3 paragraphs evaluating runtime isolation and stability)
  2. Rubric Alignment and Fairness Evaluation (bulleted review of grading logic against criteria)
  3. Vulnerability and Privacy Audit (table listing risk, impact, and mitigation)
  4. Recommended Code Patches (refactored code blocks for identified problem areas)

Self-review

  • Ensure student privacy risks related to {{student_data_scope}} are explicitly addressed.
  • Verify all suggested code modifications preserve compatibility with {{lms_platform}}.
  • Confirm test case isolation mechanisms prevent student code from altering evaluation state.
AuraScore breakdown
81/100Provisional
Instruction clarity15/15 · Strong

Explicit role, a named task, and discrete steps the model can follow.

Context architecture12/12 · Strong

Background, inputs and variables the model needs before it starts.

Constraint engineering10/12 · Adequate

Hard boundaries — what the model must and must not do.

Output specification6/14 · Thin

A named, field-level shape for the response.

Reasoning structure10/10 · Strong

Ordered work items that force analysis before an answer.

Model compatibility10/10 · Strong

Length and structure that travel across frontier models.

Token efficiency5/10 · Thin

Signal density — instruction weight without padding.

Reusability7/7 · Strong

Documented variables so the scaffold adapts to new inputs.

Robustness5/5 · Strong

Quality bar, assumptions and behaviour when inputs are thin.

Observed performance1/5 · Thin

How much real usage the template has behind it.

developers
developers-code-review
education-research
edtech
autograder
code-review