planning_ai/eval/eval.txt

**Task:**  
You are grading **two** text summaries of a source document to determine which summary more comprehensively captures the key points within the source document.  

### **Evaluation Criteria:**  
A good summary should:  
1. **Be accurate** – It should not include information that is not present in the source document.  
2. **Be comprehensive** – It should reflect all key points in the source document without omitting important details.  
3. **Be well-grounded** – It should be based entirely on the source document without adding interpretations, opinions, or external information.  

### **Scoring System:**  
- **Score 0:** Neither summary sufficiently captures the key points.  
- **Score 1:** The first summary better reflects the content of the source document.  
- **Score 2:** The second summary better reflects the content of the source document.  
- **Score 3:** Both summaries are sufficiently accurate and comprehensive.  

### **Evaluation Process:**  
1. **Compare each summary to the source document.** Identify whether each summary includes all key points, omits critical details, or introduces extraneous information.  
2. **Assess which summary better aligns with the source document.** Determine whether one summary is significantly more accurate and comprehensive.  

---

**Source Document:**  
{document}  

**Summary 1:**  
{summary_1}  

**Summary 2:**  
{summary_2}  

**Final Score (0-3):**