Live
Serverless Video Orchestration Scales 5,000+ AI Assets on Google CloudCognition’s SWE‑2 Flash: Private Weights on an Open‑Source Base and Engineer ImpactGPU Partitioning Arrives Natively in SageMaker InferenceGitHub chat integration goes GA: practical impact on workflow and governanceAgentic Primitives Taxonomy Redefines Amazon Quick Design and OperationsCutting LLM Inference Startup Overhead with Weight Daemons and Engine SnapshotsDesigning an Autonomous SOC with Amazon Bedrock AgentCore’s Tiered Multi‑Agent PatternManaging GitHub Copilot Code Review via API: Practical Implications for EngineersServerless Video Orchestration Scales 5,000+ AI Assets on Google CloudCognition’s SWE‑2 Flash: Private Weights on an Open‑Source Base and Engineer ImpactGPU Partitioning Arrives Natively in SageMaker InferenceGitHub chat integration goes GA: practical impact on workflow and governanceAgentic Primitives Taxonomy Redefines Amazon Quick Design and OperationsCutting LLM Inference Startup Overhead with Weight Daemons and Engine SnapshotsDesigning an Autonomous SOC with Amazon Bedrock AgentCore’s Tiered Multi‑Agent PatternManaging GitHub Copilot Code Review via API: Practical Implications for Engineers
GitHub

Copilot IDE integration and session controls tighten the feedback loop for engineers

AI SummaryPowered by AI

GitHub Copilot added public‑preview support for JetBrains Air and Xcode, and introduced a set of session‑management, merge‑automation, and CLI‑dashboard features across its IDE integrations. These changes let engineers automate pull‑request handling, streamline context handling, and centralise configuration, which can reduce friction in CI/CD pipelines and raise considerations around approval automation and remote session exposure.

GitHub Copilot’s November update brings two new IDE front‑ends—JetBrains Air and Xcode’s built‑in coding assistant—while expanding pull‑request automation, session handling, and a unified CLI settings view. For AI, cloud, DevOps, and security practitioners, the additions affect how code suggestions are consumed, how automated approvals are managed, and how remote tooling is exposed.

Copilot IDE integration

Copilot is now available in JetBrains Air, letting users track multiple agent sessions and run tasks either locally or through Air Teams with the same subscription. Xcode developers can access Copilot directly from the native coding assistant; the feature is in public preview and requires Xcode 27.3 beta. VS Code continues to evolve with version 1.145, offering experimental chat view modes that let users collapse or expand reasoning and tool calls without starting a new thread. Remote‑browser support and interactive canvases appear in the Agents window (preview), and recurring tasks can be scheduled with cron‑style expressions. These IDE extensions broaden the contexts where Copilot can suggest code, run automations, and surface remote UI elements.

Agent session and workflow enhancements

Session management received several usability upgrades. Users can start a session by dropping files onto the “New” sidebar entry; CSV and TSV files render as tables with sticky headers and a source‑view toggle. Sessions can now be archived in bulk when their associated pull requests merge or close, and multiple sessions can be pinned or unpinned together. Nested chats can be separated without losing conversation history, and the My Work view now stores more than ten custom panels. A new “Agent Merge” action lets Copilot merge a pull request directly from the PR view, staying enabled while required checks or rules are pending. Reviewers can be inspected, additional reviews requested, and comment text selected for quoting or discussion with Copilot. Personal instructions for assisted approval are opt‑in and apply only to local sessions, giving teams the option to pre‑approve certain tool requests.

CLI and settings consolidation

The Copilot CLI now opens a unified /settings dashboard that aggregates plugins, MCP servers, and other configuration items, with a searchable interface. Installing the CLI and the Copilot app remain separate steps, but the consolidated view reduces context switching for engineers who manage both local and cloud‑based agents.

Related CloudNinjas coverage: AI engineering.

What This Means For Practitioners

  • Evaluate the security posture of assisted approval opt‑in, especially when automating merges that bypass manual checks.
  • Update CI/CD pipelines to incorporate the new Agent Merge capability, reducing manual merge steps.
  • Test the remote‑browser and canvas previews in controlled environments before exposing them to production workloads.
  • Leverage the unified /settings dashboard to audit plugin and server configurations across teams.
  • Monitor session archiving policies to ensure that retained context does not inadvertently expose sensitive code history.
Originally published atGitHub Changelog