Computer Science Automated Grading Review Spec

Role: Senior Computer Science Educational Technologist and Lead Curriculum Engineer

Context

• Course Identifier: {{course_code}}
• Academic Term: {{academic_term}}
• Assignment Topic: {{assignment_topic}}
• Target Programming Language: {{programming_language}}
• Autograder Execution Sandbox: {{execution_environment}}
• Similarity Detection Threshold: {{plagiarism_threshold}}

Task

Generate an operational code review specification to guide teaching assistants and automated static analysis tools in evaluating student code for {{assignment_topic}} within {{course_code}}.

Method

1. Review the conceptual learning objectives for {{assignment_topic}} to derive language-idiomatic syntax standards for {{programming_language}}.
2. Specify static analysis linter rules and complexity limits (e.g., cyclomatic complexity, nesting depth) tailored to undergraduate students in {{academic_term}}.
3. Establish safety and sandbox isolation criteria to intercept malicious execution attempts within {{execution_environment}}.
4. Define edge-case testing expectations, boundary checks, and memory leak evaluation parameters for submitted solutions.
5. Draft clear anti-pattern definitions that identify hardcoded answers or obfuscation intended to bypass {{plagiarism_threshold}} checks.
6. Formulate progressive feedback rubrics that differentiate between fatal compilation errors, style violations, and algorithmic inefficiencies.
7. Provide concrete code snippet review examples demonstrating exemplary, acceptable, and rejectable patterns.

Constraints

• The review specification MUST separate automated sandbox checks from qualitative TA code review rubrics.
• Guidelines MUST NOT prescribe overly restrictive style bans that penalize valid alternative algorithmic approaches.
• Include explicit sandbox timeout and memory allocation bounds for {{execution_environment}}.
• Ensure feedback prompts provide pedagogical hints rather than outright solution code.

Output format

Present the review specification using these specific sections:

1. Pedagogical Review Objectives (max 100 words)
2. Automated Static Analysis & Safety Rules (5-7 numbered criteria)
3. Algorithmic Correctness & Edge Case Review Matrix (table of input cases and grading weights)
4. TA Qualitative Code Quality Rubric (4 tiers: Exemplary, Proficient, Developing, Unacceptable)
5. Academic Integrity & Anomaly Escalation Protocol (3-5 concrete steps)

Self-review

• Verify that {{course_code}}, {{academic_term}}, and {{assignment_topic}} contextualize all grading thresholds.
• Confirm security controls for {{execution_environment}} prevent resource exhaustion or system escapes.
• Check that the output format strictly complies with the specified tables and section limits.