Tabine recently introduced its AI-supported Code Review Agent. This GenAI assistant helps developers check their code against their own best practices and standards. This should ultimately produce safer, higher-quality code.
Development teams often create software in their own ways, so general best practices and standards for reviewing code make little sense. Moreover, argues AI coding assistant provider Tabine, using AI only provides the most value if these algorithms understand the code in the same way as the developers themselves.
According to Tabine, this should be a thing of the past with the release of Code Review Agent. The tool allows development teams to create their own rules that check the code for quality and consistency. These rules are based on their own best practices, standards, and rules imposed by their company.
Code Review Agent functionality
Developers can input rules in general language to determine their code’s requirements. Behind the scenes, Tabine’s tool turns these instructions into a set of rules that the AI can work with. Tabine also provides a set of self-defined rules that can be integrated within the overall set as desired.
Ultimately, the Code Review Agent ensures that teams can apply these self-created rules or those from the vendor during code review, whether during a pull request or in the IDE.
For example, when a developer creates a pull request that does not correspond with the set rules, the tool flags this within the code review. Then, the Code Review Agent suggests how to fix this problem.
Better code
Ultimately, this should produce higher-quality and much safer code while allowing developers to focus on more important issues.
The Code Review Agent tool is currently available as a private preview for Tabine Enterprise customers.
Also read: AI coding tools divide developers’ opinions