add a skill to automatically add apprpraite error logging

Author: teg-atlassian

.rovodev/error-handling.md

@@ -0,0 +1,67 @@
+# Error Handling Guidelines
+
+When implementing or reviewing error handling in the codebase, follow these principles:
+
+## 1. Error Logging Requirements
+
+All `try-catch` blocks should log errors appropriately based on severity and context.
+
+## 2. Error Severity Levels
+
+Use appropriate logging levels based on the error type:
+
+- **ERROR level**: Use for unexpected errors, system failures, or critical issues that require immediate attention
+- **WARN level**: Use for user-related errors such as:
+  - 401 Unauthorized (authentication issues)
+  - 403 Forbidden (permission issues)
+  - Other client errors (4xx status codes) caused by user actions
+- **DEBUG level**: Use for expected errors that don't require user intervention or are informational
+
+## 3. Retry Loop Handling
+
+When implementing retry logic:
+
+- **Do NOT log errors** on intermediate retry attempts
+- **Only log the error** if the final retry attempt fails
+- This prevents duplicate logging and log spam for transient issues
+
+Example pattern:
+```typescript
+let lastError;
+for (let attempt = 0; attempt < maxRetries; attempt++) {
+    try {
+        // operation
+        return result;
+    } catch (error) {
+        lastError = error;
+        if (attempt === maxRetries - 1) {
+            // Log only on final failure
+            logger.error('Operation failed after retries', error);
+        }
+    }
+}
+```
+
+## 4. Dual Logging Requirement
+
+All error logging must be sent to **both**:
+- **Sentry**: For error tracking and monitoring
+- **Amplitude**: For analytics and user behavior tracking
+
+Ensure both logging calls are made for every logged error.
+
+## 5. No Duplicate Logging
+
+- Each error should only be logged **once** at the appropriate level
+- Avoid logging the same error at multiple levels (e.g., both ERROR and WARN)
+- If an error is caught and re-thrown, only log at the final catch location
+- Check if an error has already been logged before logging again
+
+## Implementation Checklist
+
+When adding error handling:
+- [ ] Choose appropriate severity level (ERROR/WARN/DEBUG)
+- [ ] If in retry loop, only log on final attempt
+- [ ] Send to both Sentry and Amplitude
+- [ ] Verify no duplicate logging in the call stack
+- [ ] Include relevant context (user action, request details, etc.)

.rovodev/prompts.yml

@@ -2,3 +2,6 @@ prompts:
   - name: bump-version
     description: Bump version
     content_file: bump-version.md
+  - name: error-handling
+    description: Error handling and logging guidelines
+    content_file: error-handling.md