GitHub Copilot has added new model selection options, code review updates, and Sentry integration in the Copilot app, according to a September 14 weekly release. Updates for administrators and new agent features in VS Code are also part of this release.
Key Points
- New model selection options allow users to choose how Copilot weighs cost, quality, and response time, with tiers for efficiency, balance, and intelligence.
- The Sentry canvas in the GitHub Copilot app enables users to review errors, stack traces, and related context from Sentry reports to facilitate code fixes.
- Copilot code review now resolves addressed comments during subsequent reviews and suggests commit messages when code suggestions are applied.
- New usage metrics for VS Code Agents window are generally available, including daily active users, session counts, and user message totals for Enterprise and organization reports.
- Copilot suggestions for repository custom properties are in public preview, assisting with value selection for properties like internet-facing status.
- Users can request a higher Copilot budget from their organization or enterprise when reaching AI credit limits, a generally available feature for Copilot Business and Copilot Enterprise plans with usage-based billing.
- Agents can now run with project tools and dependencies using local Dev Containers in the Agents window, with support rolling out gradually.
Context
According to the GitHub Changelog, these updates aim to enhance the functionality and administrative capabilities of GitHub Copilot. The new features span developer workflows from debugging and code review to resource management and agent operations.
Why It Matters
These updates provide developers and administrators with more control over GitHub Copilot's behavior, better insights into its usage, and expanded capabilities for integrating it into development workflows. The changes affect how users interact with the tool for code generation, review, and debugging, as well as how organizations manage their AI credit consumption.
What To Do
- Review the new model selection options in VS Code, Copilot CLI, or the Copilot app to adjust the balance between cost, quality, and response time.
- Explore the Sentry canvas in the GitHub Copilot app to integrate crash report analysis with code fixing workflows.
- Note the new usage metrics available in the VS Code Agents window for monitoring adoption and activity.
- For Copilot Business and Copilot Enterprise users, understand the process for requesting higher AI credit budgets.
