Metabob: Generative AI for debugging & refactoring codeMetabob is an AI-based code review tool that utilizes a combination of Graph Neural Networks (GNNs) and Generative AI (LLMs). Using the underlying techniques, Metabob can detect complex logical problems and vulnerabilities, as well as general opportunities for code quality improvements. Metabob explains its detections in natural language and can generate code recommendations to fix the detections using an integrated fine-tuned LLM. Supported Languages
AI Code Review
Features
Settings
Usage
Users can request Metabob to analyze the file they are currently working on by:
Users can confirm that the Metabob Analysis is running by seeing a loading icon next to a text "Metabob: Analyzing Document" on the bottom left corner of the VS Code editor. Once the analysis has finished, users can see problematic areas highlighted in red on the file they are currently viewing and requested analysis for from Metabob.
Users can view problem details by hovering over the red highlighted problem regions in their code and clicking "More Details"
Users can generate code recommendations to fix the detected problems by:
If users think that the generated code recommendation is not correct, they can generate another recommendation by using the "REGENERATE" -button Users can view the different generated code recommendations they have asked to generate by using the arrows below the code generated code recommendation Users can apply the generated code recommendations to their code by clicking the "APPLY" -button Data Policy Metabob deletes all data from it's problem detection model one hour after the user has made their last API call. However, as Metabob integrates with third party LLMs to generate problem descriptions and code recommendations to perform fixes, Metabob has to pass data to these models and cannot control how the data is used by the companies hosting these LLMs. |