GitHub Makes Core Agent Features Generally Available in Copilot for JetBrains IDEs
Original: Major agentic capabilities improvements in GitHub Copilot for JetBrains IDEs View original →
GitHub’s March 11, 2026 update for GitHub Copilot in JetBrains IDEs is more than a feature refresh. It marks a shift toward a configurable multi-agent development environment inside the IDE. The headline change is that custom agents, sub-agents, and plan agent are now generally available, giving teams a more structured way to split responsibilities and tailor Copilot to local workflows.
Just as important, GitHub paired those new capabilities with control surfaces. Agent hooks are now in public preview, which lets developers run custom commands at key points in an agent session. The company says supported events include userPromptSubmitted, preToolUse, postToolUse, and errorOccurred, configured through a hooks.json file in .github/hooks/. Auto-approve support for MCP also expands the scope for external tool automation.
What changed in JetBrains Copilot
- Custom agents, sub-agents, and plan agent moved to general availability.
- Agent hooks entered public preview, enabling workflow automation, policy enforcement, and external integrations during agent sessions.
- Auto-approve support for MCP was added, which should reduce friction for trusted tool workflows.
- GitHub also expanded agent instruction file support, made auto model selection generally available, and added a thinking panel plus a context window usage indicator.
The reasoning UX changes are notable on their own. GitHub says extended-reasoning models such as Codex now get a dedicated thinking panel, and Anthropic thinking budgets can be configured. Combined with the new context usage indicator, that gives developers more visibility into how much reasoning and context a session is consuming, which matters as agent workflows get longer and more stateful.
The release also includes smoother sign-in, automatic opening of the chat panel after login, better device-code handling for GitHub Enterprise users, and a set of reliability fixes around terminal output, file updates, and UI glitches. Edit mode, however, has now been marked as deprecated.
The broader takeaway is that GitHub is turning Copilot for JetBrains into something closer to an operational agent platform, not just a code-completion add-on. For teams that need automation plus governance inside a familiar IDE, hooks, MCP, and role-oriented agents together make this update materially more important than a routine plugin release.
Source: GitHub changelog
Related Articles
GitHub said on March 5, 2026 that Copilot code review now runs on an agentic tool-calling architecture and is generally available for Copilot Pro, Pro+, Business, and Enterprise. The update is designed to pull wider repository context into reviews so comments are higher-signal and less noisy.
GitHub said on March 5, 2026 that GPT-5.4 is now generally available and rolling out in GitHub Copilot. The company claims early testing showed higher success rates plus stronger logical reasoning and task execution on complex, tool-dependent developer workflows.
GitHub Copilot CLI is now generally available, bringing Copilot into the terminal for standard subscribers. GitHub paired the release with broader Copilot changes including next edit suggestions, MCP-enabled agent mode, background agents, and a higher-end Pro+ plan.
Comments (0)
No comments yet. Be the first to comment!