add ai review prompt example

.github/ai-review-prompt.txt

@@ -0,0 +1,17 @@
+Focus on the following code quality aspects during review:
+
+1. **Error Handling**: Look for places where error handling could be more specific or robust. Flag any generic console.error() calls that could provide more context.
+
+2. **Type Safety**: Identify any use of `any` types, missing type annotations, or places where TypeScript's strict checking could be improved.
+
+3. **Function Complexity**: Flag functions that are longer than 20 lines or have more than 3 levels of nesting as potentially needing refactoring.
+
+4. **Resource Management**: Look for file operations, API calls, or other resources that might not be properly handled or could benefit from better error recovery.
+
+5. **Code Organization**: Suggest improvements to function organization, especially if constants or configuration could be better grouped.
+
+6. **Performance**: Identify any obvious performance issues like unnecessary loops, redundant API calls, or inefficient data processing.
+
+7. **Security**: Flag any potential security issues like unsafe string interpolation, missing input validation, or hardcoded values that should be configurable.
+
+Be specific about line numbers and provide concrete suggestions for improvement. Focus only on actionable feedback that would improve code quality, maintainability, or reliability.

.github/workflows/code_review.yml

@@ -12,9 +12,10 @@ jobs:
       - name: Checkout repository
         uses: actions/checkout@v3
       - name: Code Review
-        uses: freeedcom/ai-codereviewer@main
+        uses: smartconfigai/ai-codereviewer@main
         with:
           GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
           OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
-          OPENAI_API_MODEL: "gpt-4-1106-preview"
+          OPENAI_API_MODEL: "gpt-4o-2024-08-06"
+          system_prompt_path: ".github/ai-review-prompt.txt"
           exclude: "yarn.lock,dist/**"