49/ 50
＋ Clearly covers intent, impact, and relevant tests.
－ Some defensive checks extend slightly beyond the change scope.
Developer Work Analysis from Git Diff and Commit Message
a Code Review Expert. You are an experienced software developer with expertise in code analysis and version control systems.
| Category | Development › Coding |
|---|---|
| Tags | AnalyzingReviewingDeveloperCode |
Act as a Code Review Expert. You are an experienced software developer with expertise in code analysis and version control systems.
Your task is to analyze a developer's work based on the provided git diff file and commit message. You will:
- Assess the scope and impact of the changes.
- Identify any potential issues or improvements.
- Summarize the key modifications and their implications.
Rules:
- Focus on clarity and conciseness.
- Highlight significant changes with explanations.
- Use code-specific terminology where applicable.
Example:
Input:
- Git Diff: ${sample_diff_content}
- Commit Message: ${sample_commit_message}
Output:
- Summary: ${concise_summary_of_the_changes}
- Key Changes: ${list_of_significant_changes}
- Recommendations: ${suggestions_for_improvement}What this prompt does
This code review role is useful for reviewing code changes or turning them into a work report. It asks for clear, concise analysis and structures the output as summary, key changes, and recommendations.
Model comparison
ChatGPT has the best balance of accuracy, concision, and structure. Gemini is specific but somewhat verbose, while [C] is missing.
45/ 50
＋ Explains the logic and encapsulation improvements concretely.
－ English headings and expansive design advice reduce concision.
| Criterion | ChatGPT | Gemini | Leader |
|---|---|---|---|
| Instruction following | 10 | 9 | ChatGPT +11% |
| Accuracy | 10 | 9 | ChatGPT +11% |
| Specificity | 10 | 10 | Tie |
| Structure | 10 | 9 | ChatGPT +11% |
| Right length | 9 | 8 | ChatGPT +13% |
Scored 1–10 by gpt-5.6-sol with model names hidden (2026-09-25). This is an AI review, not a measurement.
Read full answers
We gave three models the same input and copied their answers unedited. Each ran in its CLI (an agent harness), and answers in the ChatGPT or Claude apps or on the web may differ. Outputs are in Korean.
[Pasted material]
Git Diff:
diff --git a/src/main/java/com/shopmate/payment/PaymentService.java b/src/main/java/com/shopmate/payment/PaymentService.java
index 3a1f2c9..8b7e421 100644
--- a/src/main/java/com/shopmate/payment/PaymentService.java
+++ b/src/main/java/com/shopmate/payment/PaymentService.java
@@ -42,7 +42,10 @@ public class PaymentService {
public PaymentResult processPayment(Order order) {
- BigDecimal amount = order.getDiscount().getFinalAmount();
+ Discount discount = order.getDiscount();
+ BigDecimal amount = (discount != null)
+ ? discount.getFinalAmount()
+ : order.getTotalAmount();
return pgClient.charge(order.getUserId(), amount);
}
Commit Message:
fix: 쿠폰 미적용 주문에서 결제 시 NPE 발생하는 문제 수정
order.getDiscount()가 null인 경우(쿠폰 미사용 주문) getFinalAmount() 호출 시
NullPointerException이 발생하던 버그 수정. discount가 없으면 정가(totalAmount)로 결제하도록 처리.요약: 쿠폰이 적용되지 않은 주문에서
order.getDiscount()가null일 때 발생하던NullPointerException을 방지하고, 할인 정보가 없으면 주문 총액(totalAmount)을 결제하도록 수정했습니다.주요 변경사항:
Discount객체를 지역 변수로 추출했습니다.- 할인 정보가 있으면
getFinalAmount()를 사용합니다. - 할인 정보가 없으면
order.getTotalAmount()를 사용합니다. - 결제 승인 로직의 NPE 발생 경로가 제거되었습니다.
영향 및 검토 의견:
- 쿠폰 미사용 주문도 정상적으로 정가 결제가 가능해져 버그 수정 의도와 일치합니다.
order자체 또는order.getTotalAmount()가null일 가능성이 있다면 추가적인 검증이 필요합니다.- 할인 금액 계산 규칙이 다른 곳에서도 사용된다면, 결제 서비스에서 직접 분기하기보다
Order또는 금액 계산 도메인 로직으로 캡슐화하는 것을 고려할 수 있습니다.
권장사항:
- 할인 적용 주문, 쿠폰 미사용 주문,
totalAmount경계값에 대한 단위 테스트를 추가하세요. - 결제 요청 직전 금액이 음수이거나
null인 경우를 검증하는 방어 로직이 있는지 확인하세요.
- 할인 적용 주문, 쿠폰 미사용 주문,
More in this category
| 12-Month AI and Computer Vision Roadmap for Defense Applications | |
| 2046 Puzzle Game Challenge | |
| 21st.dev component prompt | |
| 3D FACTORY | |
| 3D FPS Game |