FAQ & Troubleshooting

Q: What are the AI models provided by DeputyDev?

A: DeputyDev uses OpenAI's GPT-4 and Anthropic's Claude 3.5 Sonnet as the default models due to their superior performance in code evaluation tasks. Both models have achieved the highest scores on the HumanEval and MMLU benchmarks, making them the top choices for ensuring high-quality code reviews.

Q: How do I make queries?

A: Use the #deputydev or #dd tag followed by your question or statement to initiate context-aware queries. Ensure that your query is clear and specific.

Q: How do I fix context-aware queries not responding?

A: Make sure you are using the #deputydev tag to initiate queries. Check your internet connection.

Q: What can I do if business logic alignment checks are failing?

A: Ensure that the user stories are well-defined and included in the PR descriptions. Verify that the business logic rules are correctly set up in the configuration settings. Re-run the alignment checks after making any necessary adjustments.

Q: How can I provide feedback on DeputyDev's comments?

A: You can use the feedback feature to leave responses in the form of `#like`, `#dislike`, or `#suggestion`. If you like the comment, respond with a #like followed by any specifics (if applicable). If you don't like the comment, respond with a #dislike followed by any specifics (if applicable). If you're neutral and simply want to leave a suggestion, respond with a #suggestion followed by the suggestion.