Third-party tools are added to Google's enterprise-focused coding assistance, Code Assist

On Tuesday, Google announced that its AI-powered code completion service, Gemini Code Assist, now supports third-party tools.

Initially introduced in April as a rebranded version of its earlier Duet AI service, Code Assist is designed for enterprise developers and integrates with popular development environments like VS Code and JetBrains via plug-ins. Powered by Google’s Gemini AI models, it enables advanced reasoning and modifications across substantial portions of code.

The newly added tools, currently in private preview, allow Code Assist to incorporate real-time data and external application information into the development process. Google aims to streamline workflows by integrating multiple technologies directly into the coding environment, reducing distractions. This feature was detailed in a blog post by Ryan Salva, Google’s director of product management, and Prithpal Bhogill, a group product manager.

“This tools feature eliminates the friction caused by context switching,” Salva and Bhogill explained. “Developing scalable, secure applications goes beyond writing excellent code — it requires solutions for productivity, observability, security, databases, and more.

At present, only Google Cloud partners can develop tools for Code Assist.

According to Salva and Bhogill, “Tools allow developers to interact with or retrieve data from various parts of their engineering systems, particularly services outside their development environment. For instance, you could summarize Jira issue comments, identify the last contributor to a file in Git, or highlight recent live site issues from Sentry.”

At launch, Code Assist includes tools from GitLab, GitHub, Sentry.io, Atlassian Rovo, Snyk, and Google Docs. Google Cloud partners interested in creating additional tools are encouraged to contact their partner managers.