Skip to content

Latest commit

 

History

History
79 lines (57 loc) · 7.16 KB

File metadata and controls

79 lines (57 loc) · 7.16 KB
title Using GitHub Copilot in JetBrains IDEs
shortTitle Copilot in JetBrains
intro Learn about the different ways to use {% data variables.product.prodname_copilot %} in {% data variables.product.prodname_jetbrains_ides %}, including the {% data variables.product.prodname_copilot %} plugin, JetBrains AI Assistant, and {% data variables.copilot.copilot_cli_short %}.
versions
feature
copilot
contentType concepts
category
Learn about Copilot
allowTitleToDifferFromFilename true

Introduction

There are three ways to use {% data variables.product.prodname_copilot %} in {% data variables.product.prodname_jetbrains_ides %}: the {% data variables.product.prodname_copilot %} plugin, {% data variables.product.prodname_copilot %} as an agent in JetBrains AI Assistant, and {% data variables.copilot.copilot_cli %} in the integrated terminal. Each entry point provides a different set of capabilities depending on how you prefer to work.

Comparing entry points

{% data variables.product.prodname_copilot %} plugin {% data variables.product.prodname_copilot %} in AI Assistant {% data variables.copilot.copilot_cli_short %}
Best for Comprehensive AI coding workflow Quick {% data variables.product.prodname_copilot_short %} access without a separate plugin Terminal-first workflows
Entry point Chat panel, inline chat, code suggestions, code review, commit message Default agent picker, ACP Registry Terminal or shell
Code completion Yes Not included Not included
Next edit suggestions Yes Coming soon Not included
Agentic experience Multiple agent harnesses {% data variables.product.prodname_copilot_short %} as agent via ACP, Default agent picker {% data variables.copilot.copilot_cli_short %}
IDE tools Yes Yes Not included
Model selection Yes Yes Yes
Inline chat Yes Not included Not included
Code review Yes Not included Not included
Commit message generation Yes Not included Not included
Subscription {% data variables.product.prodname_copilot %} {% data variables.product.prodname_copilot %} {% data variables.product.prodname_copilot %}

{% data variables.product.prodname_copilot %} plugin

The {% data variables.product.prodname_copilot %} plugin for {% data variables.product.prodname_jetbrains_ides %} is the most comprehensive way to use {% data variables.product.prodname_copilot_short %} and is the recommended choice.

The plugin is transitioning from its local agent harness to {% data variables.copilot.copilot_cli_short %} as the default agent harness, which brings faster feature parity and higher-quality results. For more information, see {% data variables.copilot.copilot_cli_short %} is becoming the default agent harness in {% data variables.product.prodname_copilot %} for JetBrains. For installation instructions, see AUTOTITLE.

  • Code completion and next edit suggestions: {% data variables.product.prodname_copilot_short %} suggests completions as you type and proactively predicts your next intended edit.
  • Multiple agent harnesses: The plugin ships its own agent experience and partners with other agent providers, giving you multiple interaction modes.
  • Full model and feature support: All {% data variables.copilot.copilot_chat_short %} models, code completion modes, and bring-your-own-key features are available as they are released.
  • Inline chat: Explain, refactor, document, or generate code directly in the editor gutter, without switching to a separate panel.
  • Code review: {% data variables.product.prodname_copilot_short %} analyzes your changes and surfaces actionable feedback, flagging potential bugs, style violations, and logic issues.
  • Commit message generation: {% data variables.product.prodname_copilot_short %} inspects your staged changes and generates a clear, conventional commit message.

{% data variables.product.prodname_copilot %} in JetBrains AI Assistant

Note

{% data variables.product.prodname_copilot %} in AI Assistant provides chat and agent capabilities only. It does not include code completion, next edit suggestions, inline chat, code review, or commit message generation.

{% data variables.product.prodname_copilot %} is available as a native agent in JetBrains AI Assistant through the Agent Client Protocol (ACP). The ACP is an open standard for connecting AI agents to the IDE. If you have a valid {% data variables.product.prodname_copilot_short %} subscription, {% data variables.product.prodname_copilot_short %} appears in the AI Assistant agent picker automatically.

This integration is designed for developers who prefer to work inside the AI Assistant chat panel or who want {% data variables.product.prodname_copilot_short %} available without installing an additional plugin.

  • No updates required: The {% data variables.product.prodname_copilot_short %} agent is bundled directly with AI Assistant and kept current automatically. No separate plugin to install, update, or maintain.
  • Chat-centric workflow: Ideal for multi-step reasoning tasks—describe a goal, let {% data variables.product.prodname_copilot_short %} plan and propose changes, and iterate conversationally.
  • Model selection: Switch {% data variables.product.prodname_copilot_short %} models or adjust reasoning depth without leaving the chat panel.

Using {% data variables.product.prodname_copilot %} in AI Assistant

  1. Open JetBrains AI Assistant by pressing Alt+A (Windows/Linux) or Command+Shift+A (macOS), or click the AI Assistant icon in the right tool window.
  2. In the agent picker dropdown at the top of the chat panel, select {% data variables.product.prodname_copilot %}.
  3. Enter a prompt and start chatting.

The ACP Registry

The ACP Registry is the catalog of agents that AI Assistant knows about. When the IDE starts, it consults the registry to discover which agents are available. {% data variables.product.prodname_copilot %}'s ACP entry is part of the default registry, so {% data variables.product.prodname_copilot_short %} appears in your agent list automatically when you have a valid subscription and the required credentials in place.

For more information about ACP, see the ACP documentation. For technical details on running {% data variables.copilot.copilot_cli_short %} as an ACP server, see AUTOTITLE.

{% data variables.copilot.copilot_cli %} in the integrated terminal

{% data variables.copilot.copilot_cli %} brings {% data variables.product.prodname_copilot_short %}'s capabilities directly to the terminal. It is optimized for command-line workflows and can run on macOS, Linux, or Windows.

Further reading