GitHub Copilot for JetBrains adds enterprise model defaults and MCP startup controls

GitHub released a Copilot for JetBrains update on October 10 that gives enterprise teams more control over how AI coding assistance starts inside JetBrains IDEs. The change is not a new model launch. It is a management and workflow update for teams that already depend on Copilot in IntelliJ IDEA, PyCharm, WebStorm and related JetBrains tools.

The most important admin feature is managed default model selection. GitHub says enterprise administrators can now choose any available Copilot agent model as the default for new conversations, while still leaving users an explicit model picker. That matters for organizations trying to standardize behavior across teams without removing every developer choice.

Why developers should care

The update also adds a diagnostic Fix action. When a diagnostic intention menu identifies a problem, developers can open inline chat and ask Copilot to repair it. GitHub says the action uses agent mode when available and ask mode otherwise, which should make the feature useful across mixed setups where not every workspace has the same agent capabilities enabled.

A second control targets Model Context Protocol usage. GitHub added an MCP setting that can disable automatic server startup for Copilot and Claude. For teams experimenting with tool-connected agents, that is a practical security and reliability lever. It gives administrators a way to prevent configured tools from becoming active simply because a coding assistant session starts.

• Enterprise admins get default model management for new Copilot conversations.
• Developers get a faster path from IDE diagnostics to proposed fixes.
• Teams can stop automatic MCP server startup for Copilot and Claude.
• GitHub has ended support for JetBrains IDE 2025.1, so the plugin now requires JetBrains IDE 2025.2 or later.

The CyberOGZ read

This update is a reminder that AI coding tools are becoming managed workplace infrastructure, not just editor plugins. The useful decision point for teams is whether model defaults and MCP startup rules are documented before rollout. If they are left to individual machines, agent behavior can drift across projects. If they are governed too tightly, developers may lose the flexibility that made the tool valuable in the first place.

For JetBrains shops, the immediate action is simple: check IDE versions, review Copilot managed settings, and decide whether automatic MCP startup matches internal policy. The headline feature is control, but the operational burden is making those controls visible before developers run into them mid-task.

Sources

• github.blog

Cover photo by Paras Katwal on Pexels, used under the Pexels License.