Skip to main content

GitHub Copilot cloud agent

Find out how Copilot can research a repository, plan and make code changes, and create pull requests for you to review.

These articles provide how-to information for using Copilot cloud agent. For an overview of Copilot cloud agent, see the articles under Concepts for GitHub Copilot cloud agent.

Research, plan, and iterate on code changes with Copilot cloud agent

Perform deep research on a repository, create an implementation plan, and make iterative code changes on a branch, then create a pull request when you're ready.

Asking GitHub Copilot to create a pull request

You can ask Copilot to create a pull request from many places, including GitHub Issues, the agents panel, Copilot Chat, the GitHub CLI, and agentic coding tools and IDEs with Model Context Protocol (MCP) support.

Asking GitHub Copilot to make changes to an existing pull request

You can ask Copilot to make changes to an existing pull request by mentioning @copilot.

Provide visual inputs to GitHub Copilot

You can attach images to issues or Copilot Chat prompts to help Copilot understand your task.

Tracking GitHub Copilot's sessions

You can use the agents panel or page, Visual Studio Code, JetBrains IDEs, Eclipse, the GitHub CLI, Raycast and session logs to track Copilot's progress and understand its approach.

Reviewing a pull request created by GitHub Copilot

After Copilot creates a pull request, you should review it. You can mention @copilot to ask Copilot to make changes, or push changes yourself.

Creating custom agents for Copilot cloud agent

You can create specialized agents with tailored expertise for specific development tasks.

Testing and releasing custom agents in your organization or enterprise

Ensure your custom agents are performant and compliant before releasing them to your company.

Creating agent skills for GitHub Copilot

You can modify Copilot's behavior and abilities when it works on particular tasks.

Extending GitHub Copilot cloud agent with the Model Context Protocol (MCP)

Learn how to use the Model Context Protocol (MCP) to extend the capabilities of Copilot cloud agent.

Integrating Copilot cloud agent with Jira

You can use the GitHub integration in Jira to provide context and open pull requests, all from within your Jira workspace.

Integrating Copilot cloud agent with Slack

Provide context to the Copilot cloud agent and open pull requests, all from within your Slack workspace.

Integrating Copilot cloud agent with Teams

You can use the GitHub integration in Teams to provide context and open pull requests all from within your Teams channels.

Integrating Copilot cloud agent with Linear

Use the Copilot integration in Linear to provide context and open pull requests, all from within your Linear workspace.

Integrating Copilot cloud agent with Azure Boards

Use the Copilot integration in Azure Boards to send work items directly to Copilot cloud agent and generate pull requests, all from within your Azure DevOps workspace.

Changing the AI model for GitHub Copilot cloud agent

In supported entrypoints, when starting a task with Copilot cloud agent, you can select the model used.

Configuring settings for GitHub Copilot cloud agent

Learn how to configure settings for Copilot cloud agent

Customizing the development environment for GitHub Copilot cloud agent

Learn how to customize GitHub Copilot's development environment with additional tools.

Customizing or disabling the firewall for GitHub Copilot cloud agent

Learn how to control the domains and URLs that Copilot cloud agent can access.

Using hooks with GitHub Copilot agents

Extend and customize GitHub Copilot agent behavior by executing custom shell commands at key points during agent execution.

Troubleshooting GitHub Copilot cloud agent

Learn how to resolve problems that may occur when you assign tasks to Copilot.