1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
|
---
description: When stuck in debugging loops, break the cycle by minimizing to an MVP, removing debugging cruft, and documenting the issue completely for a fresh approach
globs: *.py
alwaysApply: true
---
# Avoid Debug Loops
When debugging becomes circular and unproductive, follow these steps:
## Detection
- You have made multiple unsuccessful attempts to fix the same issue
- You are adding increasingly complex code to address errors
- Each fix creates new errors in a cascading pattern
- You are uncertain about the root cause after 2-3 iterations
## Action Plan
1. **Pause and acknowledge the loop**
- Explicitly state that you are in a potential debug loop
- Review what approaches have been tried and failed
2. **Minimize to MVP**
- Remove all debugging cruft and experimental code
- Revert to the simplest version that demonstrates the issue
- Focus on isolating the core problem without added complexity
3. **Comprehensive Documentation**
- Provide a clear summary of the issue
- Include minimal but complete code examples that reproduce the problem
- Document exact error messages and unexpected behaviors
- Explain your current understanding of potential causes
4. **Format for Portability**
- Present the problem in quadruple backticks for easy copying:
````
# Problem Summary
[Concise explanation of the issue]
## Minimal Reproduction Code
```python
# Minimal code example that reproduces the issue
```
## Error/Unexpected Output
```
[Exact error messages or unexpected output]
```
## Failed Approaches
[Brief summary of approaches already tried]
## Suspected Cause
[Your current hypothesis about what might be causing the issue]
````
This format enables the user to easily copy the entire problem statement into a fresh conversation for a clean-slate approach.
|
