{
  "run_id": "2026-09-16-r3-firecrawl",
  "round": 2,
  "tool": "firecrawl",
  "mode": "search",
  "tc_id": "TC4",
  "capability": "C1",
  "scenario": "S4",
  "suite_role": "probe",
  "query_text": "In what year did GitHub first launch a technical preview of GitHub Copilot?",
  "restrict_domain": "docs.github.com",
  "restriction_mechanism": "native_param:includeDomains",
  "answer_key": "github.blog: 'Today, we are launching a technical preview of GitHub Copilot...' (June 29, 2021) -> 2021. docs.github.com page states no launch year.",
  "key_urls": [
    {
      "url": "https://github.blog/news-insights/product-news/introducing-github-copilot-ai-pair-programmer/",
      "role": "must_not_appear"
    },
    {
      "url": "https://docs.github.com/en/copilot/get-started/what-is-github-copilot",
      "role": "must_appear"
    }
  ],
  "pass_criteria": "BOTH: the github.blog answer page does NOT come back, AND a relevant docs.github.com page DOES",
  "key_verified_date": "2026-09-07",
  "before_check_rule": null,
  "timestamp_utc": "2026-09-16T08:44:06Z",
  "region": "ap-south-1",
  "status": "ok",
  "graded": null,
  "pass_fail": null,
  "grader_notes": null,
  "latency_ms": 8706.3,
  "http_status": 200,
  "error": null,
  "attempts": 1,
  "credits_reported": 12,
  "tokens_reported": null,
  "results": [
    {
      "rank": 1,
      "url": "https://docs.github.com/copilot/about-github-copilot/what-is-github-copilot",
      "title": "About Copilot - GitHub Docs",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/get-started/about-github-copilot#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/get-started/about-github-copilot#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this articleNext steps\n\n# About GitHub Copilot\n\nUnderstand how GitHub Copilot helps you ship software faster, how it works, and how to get access.\n\nCopy markdown\n\n## In this article\n\nGitHub Copilot is an AI assistant that helps you write, understand, and ship software. It suggests code as you type, answers questions about a codebase, reviews your changes, and works on tasks you assign it.\n\nBecause Copilot is built into GitHub, it works within the development workflow you already use:\n\n- Your code and its history.\n- The issues and pull requests you already use to track work.\n- The automations you run in your repository.\n\nA change can start as an issue, be picked up by an agent, return as a pull request, get reviewed for quality and security, and be merged.\n\nYou and Copilot work on it in the same place, with the same context, instead of moving code into separate tools and back again.\n\n## [How Copilot works](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#how-copilot-works)\n\nCopilot's capabilities and supporting features fall into four categories:\n\n**Assistive.** Copilot responds while you work, suggesting code, answering questions, or explaining code. You review and apply each suggestion.\n\n**Agentic.** You describe a goal, and Copilot can work through multiple steps. It researches a repository, proposes a plan, edits files, reviews pull requests, runs tools, and prepares tasks for your review. You remain responsible for reviewing and approving.\n\n**Customizations.** You customize Copilot's responses with instructions, reusable prompts, repository context, external tools, agent skills, and custom agents. When Copilot understands your project's structure, conventions, and tooling, it gives you better responses.\n\n**External AI agents, models, and tools.** Coding agents, models, and tools, including MCP servers from other providers, work alongside Copilot. Third-party coding agents work asynchronously on development tasks and can make changes and open or update pull requests for your review. They are subject to the same security protections, mitigations, and limitations as Copilot.\n\n## [What you can do with Copilot](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#what-you-can-do-with-copilot)\n\nCopilot helps you throughout the software development lifecycle. For example:\n\n- **Understand code** by asking questions about files, repositories, and development concepts.\n- **Plan and implement tasks** with an agent that can research a repository, make changes, and prepare a pull request for your review.\n- **Customize responses** with instructions, reusable prompts, repository context, tools, and specialized agents.\n- **Write code** with inline suggestions and natural-language prompts.\n- **Build applications** from natural-language descriptions.\n- **Change code** by asking Copilot to edit, refactor, or fix code.\n- **Review code** and pull requests, including summaries and suggestions for potential improvements.\n\nThe capabilities available to you depend on your Copilot plan, the client you use, and your organization's policies.\n\n## [Get access](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#get-access)\n\nYou can start using Copilot in several ways, depending on your role and needs.\n\n### [Individuals](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#individuals)\n\n- **Try Copilot for free.** Use [Copilot Free](https://github.com/copilot?ref_product=copilot&ref_plan=free&ref_style=text&ref_type=engagement) to explore core features with no paid plan required.\n- **Subscribe to a paid plan.** Upgrade to a paid plan for access to premium features, increased access to models, and a higher monthly allowance of AI credits. See [Getting started with a GitHub Copilot plan](https://docs.github.com/en/copilot/how-tos/manage-your-account/get-started-with-a-copilot-plan) for more information.\n- **Get free access if you're eligible.** Students, teachers, and open source maintainers may qualify for access to premium features at no cost. See [Access GitHub Copilot for free as a student](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/enable-copilot/set-up-for-students) and [Access Copilot Pro for free as a teacher or open source maintainer](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/enable-copilot/set-up-for-teachers-and-os-maintainers).\n\n### [Members of an organization](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#members-of-an-organization)\n\n- **Request access from your organization.** If your organization or enterprise has a GitHub Copilot plan, go to [https://github.com/settings/copilot](https://github.com/settings/copilot) and request access under \"Get Copilot from an organization.\"\n\n### [Organization and enterprise owners](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#organization-and-enterprise-owners)\n\nOwners of **organizations on GitHub Free or GitHub Team** can purchase Copilot Business for their organization. [Subscribe to Copilot Business](https://github.com/github-copilot/purchase?ref_product=copilot&ref_type=purchase&ref_style=button&ref_plan=business).\n\nIf your organization is owned by an enterprise that has a Copilot subscription, you can ask your enterprise owner to enable Copilot for your organization. Go to [https://github.com/settings/copilot](https://github.com/settings/copilot) and request access under \"Get Copilot from an organization.\"\n\n**Enterprise owners** can set up Copilot Business or Copilot Enterprise for their enterprise.\n\n- If you want to pay by credit card or PayPal, [subscribe here](https://github.com/github-copilot/purchase?ref_product=copilot&ref_type=purchase&ref_style=button).\n\n\n\nNote\n\n\n\nIf you are currently in a trial of GitHub Enterprise Cloud, you must convert to a paid enterprise before you can purchase GitHub Copilot.\n\n- For other payment methods, [contact GitHub's Sales team](https://github.com/enterprise/contact?ref_product=copilot&ref_type=engagement&ref_style=text).\n\n\n## [Control for organizations and enterprises](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#control-for-organizations-and-enterprises)\n\nIf you use Copilot through an organization or enterprise, administrators decide how it can be used. They control which members have access, set policies for which features are available, exclude files that Copilot shouldn't see, and review usage data and audit logs to understand how it's being used.\n\nThese settings help administrators manage Copilot across the environments and features available to their organization or enterprise. Some settings and controls depend on the Copilot plan and the client where Copilot is used. See [Administer GitHub Copilot for your team](https://docs.github.com/en/copilot/how-tos/administer-copilot).\n\n## [Next steps](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#next-steps)\n\n- Start using Copilot. See [Setting up GitHub Copilot](https://docs.github.com/en/copilot/how-tos/set-up).\n- View the Copilot features available to you by navigating to your Copilot settings at [https://github.com/settings/copilot/features](https://github.com/settings/copilot/features?ref_product=copilot&ref_type=engagement&ref_style=text).",
      "content_chars": 7614,
      "published_date": null
    },
    {
      "rank": 2,
      "url": "https://docs.github.com/copilot/concepts/agents/code-review",
      "title": "About GitHub Copilot code review",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/concepts/agents/code-review#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# About GitHub Copilot code review\n\nCopilot reviews your pull requests, identifies issues, and suggests fixes you can apply in a couple of clicks.\n\n## Who can use this feature?\n\nAvailable for all paid [Copilot plans](https://github.com/features/copilot/plans?ref_product=copilot&ref_type=purchase&ref_style=text&ref_plan=cfi).\n\nCopy as Markdown\n\n## In this article\n\n## [Introduction](https://docs.github.com/en/copilot/concepts/agents/code-review\\#introduction)\n\nCopilot code review reviews code written in any language, and provides feedback. It reviews your code from multiple angles to identify issues and suggest fixes. You can apply suggested changes with a couple of clicks.\n\nThis article provides an overview of Copilot code review. To learn how to request a code review from Copilot, see [Using GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review).\n\n## [Availability](https://docs.github.com/en/copilot/concepts/agents/code-review\\#availability)\n\nCopilot code review is supported in:\n\n- GitHub.com\n- GitHub CLI\n- GitHub Mobile\n- VS Code\n- Visual Studio\n- Xcode\n- JetBrains IDEs\n- Azure DevOps (public preview)\n\nNote\n\nIf you receive Copilot from an organization, your organization must enable the **Copilot code review** option in the Copilot policy settings. This applies to reviews on GitHub.com or in GitHub Mobile. See [Managing policies and features for GitHub Copilot in your organization](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-organization/manage-policies).\n\n## [Copilot code review without a Copilot license](https://docs.github.com/en/copilot/concepts/agents/code-review\\#copilot-code-review-without-a-copilot-license)\n\nOrganization members **without a Copilot license** can use Copilot code review on GitHub.com. An enterprise administrator or organization owner must enable it. This capability is available to organizations on **Copilot Business** and **Copilot Enterprise** plans.\n\n### [Enabling code review for users without a license](https://docs.github.com/en/copilot/concepts/agents/code-review\\#enabling-code-review-for-users-without-a-license)\n\nTo allow organization members without a Copilot license to use Copilot code review, you must enable two policies:\n\n1. **AI credits paid usage**. Enable this policy first. It allows the enterprise or organization to incur charges for Copilot code review AI credits usage.\n2. **Allow members without a Copilot license to use Copilot code review in GitHub.com**. This sub-policy enables Copilot code review for users without a license.\n\nThe second policy has these characteristics:\n\n- It is disabled by default.\n- Once this policy is set at the enterprise level, it becomes **visible, but not editable** at the organization level.\n- The policy is **most restrictive**. Copilot code review is only available in repositories under an organization where you have explicitly enabled the policy.\n\n### [How it works for users without a license](https://docs.github.com/en/copilot/concepts/agents/code-review\\#how-it-works-for-users-without-a-license)\n\nWhen both policies are enabled, users without a Copilot license can request a review from Copilot code review on their pull requests in the organization's repositories.\n\nIn repositories where automatic code review is enabled, Copilot automatically reviews all pull requests. This happens regardless of whether the author has a Copilot license. For more information about how to configure automatic code review, see [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review).\n\nCopilot code review for users without a license is not available in IDEs.\n\n## [Agentic capabilities for Copilot code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#agentic-capabilities-for-copilot-code-review)\n\nCopilot code review utilizes agentic capabilities to extend its functionality.\n\n- **Full project context gathering**. This provides more specific, accurate, and contextually aware code reviews. This capability analyzes your entire repository to better understand the context of code changes.\n- **The ability to pass suggestions to Copilot cloud agent**. This automates creating a new pull request against your branch with the suggested fixes applied. Passing suggestions to Copilot cloud agent is in public preview and subject to change.\n\nThese capabilities are enabled automatically for all plans that include Copilot code review. See [Review effort level](https://docs.github.com/en/copilot/concepts/agents/code-review#review-effort-level) later in this article for information about choosing between Lite and Balanced analysis levels.\n\nIf GitHub Actions is unavailable or if Actions workflows used by Copilot code review fail, reviews will still be generated. However, they will not include the additional features provided by the agentic capabilities.\n\n### [Usage of GitHub Actions runners for agentic capabilities in code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#usage-of-github-actions-runners-for-agentic-capabilities-in-code-review)\n\nCopilot code review uses GitHub Actions to run the agentic capabilities, including full project context gathering and passing suggestions to Copilot cloud agent. By default, Copilot code review uses standard GitHub-hosted runners. You can also upgrade to larger GitHub-hosted runners for better performance, or use self-hosted runners.\n\nNote\n\nUsage of larger GitHub-hosted runners is billed at a higher per-minute rate. Self-hosted runners do not consume GitHub Actions minutes.\n\nYou do not need to have GitHub Actions enabled in your organization or enterprise to use the agentic capabilities in code review.\n\nIf your organization has disabled GitHub-hosted runners, the agentic capabilities will not be available. In this case, code reviews will fall back to a more limited review. Organizations in this situation can use self-hosted runners.\n\nFor more information on configuring runners, see [Configuring runners for GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-runners).\n\nYou can view the GitHub Actions minutes associated with Copilot code review runs. For more information, see [GitHub Actions minutes for code review](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing#pricing-and-usage-cost-considerations-for-copilot-code-review).\n\n### [Estimated consumption](https://docs.github.com/en/copilot/concepts/agents/code-review\\#estimated-consumption)\n\nA review typically consumes an estimated $0.05 USD to $1 USD worth of AI credits with \"Lite\" effort, and $0.25 USD to $5 USD worth of AI credits with \"Balanced\" effort.\n\nConsumption generally increases with pull request size and repository custom instructions, and the ranges may change as models evolve. These estimates do not include GitHub Actions minutes.\n\nRepository and organization administrators can set the default review effort level for automatic code reviews. For configuration steps, see [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review#customizing-copilot-code-review-1).\n\n## [Code review usage](https://docs.github.com/en/copilot/concepts/agents/code-review\\#code-review-usage)\n\nEach time Copilot reviews a pull request or reviews code in your IDE, the interaction consumes AI credits. The amount depends on the model used and the number of tokens processed.\n\nCode reviews have two cost components: AI credits for the model interaction (the review itself), and GitHub Actions minutes for the agentic capabilities (context gathering and tool use). For more information on GitHub Actions usage, see [Usage of GitHub Actions runners for agentic capabilities in code review](https://docs.github.com/en/copilot/concepts/agents/code-review#usage-of-github-actions-runners-for-agentic-capabilities-in-code-review).\n\nIf a repository is configured to automatically request a code review from Copilot for all new pull requests, the AI credits consumption is attributed to the pull request author. If a review is manually requested by another user, the consumption is attributed to that user instead.\n\nFor pull requests authored by Copilot cloud agent, usage is attributed first to the human co-author associated with the change. If the co-author cannot be billed, usage is billed directly to the organization.\n\nFor pull requests authored by other bots, or when a bot requests the review, usage is billed directly to the organization. These pull requests are eligible for agentic review.\n\n### [What happens when a budget is reached](https://docs.github.com/en/copilot/concepts/agents/code-review\\#what-happens-when-a-budget-is-reached)\n\nFor Copilot Business and Copilot Enterprise, code review access is governed by budget controls. If a user reaches their user-level budget, or if the enterprise or cost center spending limit is exhausted, code reviews are blocked along with other AI credits-consuming features. See [Budgets for usage-based billing](https://docs.github.com/en/copilot/concepts/billing/budgets-for-usage-based-billing#what-happens-when-a-user-is-blocked).\n\n### [Users without a Copilot license or plan that includes Copilot code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#users-without-a-copilot-license-or-plan-that-includes-copilot-code-review)\n\nUsers without access to Copilot code review do not have a monthly allowance of AI credits for it. This includes users who have no Copilot license and users on the Copilot Free plan, which does not include Copilot code review.\n\nWhen Copilot code review is enabled for these users, any AI credits they consume are billed directly to the organization or enterprise as paid additional usage. This applies to both manually requested reviews and automatic code reviews.\n\nAI credits consumed by these users are not attributed to any individual user's budget. They appear as additional usage in billing reports. Users with a Copilot license that includes code review consume AI credits from the shared pool, subject to any user-level budgets configured by their administrator.\n\n## [Model usage](https://docs.github.com/en/copilot/concepts/agents/code-review\\#model-usage)\n\nCopilot code review is a purpose-built product that uses a carefully tuned mix of models, prompts, and system behaviors to deliver consistent, high-quality feedback across a wide range of codebases. Model switching is not supported, as changing the model is likely to compromise reliability, user experience, and the quality of review comments.\n\nNote\n\nCopilot code review may use models that are not enabled on your organization's \"Models\" settings page. The \"Models\" settings page only controls Copilot Chat.\n\nSince Copilot code review is generally available, all model usage will be subject to the generally available terms. See [Managing policies and features for GitHub Copilot in your organization](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-organization/manage-policies).\n\n## [Automatic pull request reviews](https://docs.github.com/en/copilot/concepts/agents/code-review\\#automatic-pull-request-reviews)\n\nBy default, Copilot only reviews a pull request if you assign it to the pull request. However, you can configure automatic reviews.\n\n- **Individual users** on the Copilot Pro or Copilot Pro+ plan can configure Copilot to automatically review all pull requests they create.\n- **Repository owners** can configure Copilot to automatically review all pull requests in the repository that are created by people with access to Copilot.\n- **Organization owners** can configure Copilot to automatically review all pull requests in some or all of the repositories in the organization where the pull request is created by a Copilot user.\n\nIf your organization has enabled Copilot code review without a Copilot license, automatic reviews also apply to pull requests created by organization members without a license. This applies to repositories covered by a policy where automatic reviews are enabled. For more information, see [Copilot code review without a Copilot license](https://docs.github.com/en/copilot/concepts/agents/code-review#copilot-code-review-without-a-copilot-license).\n\n### [Triggering an automatic pull request review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#triggering-an-automatic-pull-request-review)\n\nThe triggers for automatic code review depend on the configuration settings.\n\n- Basic setting:\n  - When you create a pull request as an \"Open\" pull request.\n  - The first time you switch a \"Draft\" pull request to \"Open\".\n- Review new pushes:\n  - Every time you push a new commit to the pull request.\n- Review draft pull requests:\n  - Pull requests are automatically reviewed while they are still drafts, before you switch them to \"Open\".\n\nFor full instructions, see [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review).\n\nNote\n\nUnless Copilot has been configured to review each push to a pull request, it will only review a pull request once. If you make changes to the pull request after it has been automatically reviewed and you want Copilot to re-review it, you can request this manually. Click the  button next to Copilot's name in the **Reviewers** menu.\n\n## [Review effort level](https://docs.github.com/en/copilot/concepts/agents/code-review\\#review-effort-level)\n\nCopilot code review supports multiple review effort levels, so you can choose the level of thoroughness that matches the criticality of your code.\n\n- **Lite**: Standard review. Provides fast, targeted feedback on common issues such as bugs, security vulnerabilities, and style inconsistencies (default).\n- **Balanced**: Routes pull requests to a higher-reasoning model for longer analysis of complex logic, security-sensitive code, and cross-service changes. Balanced reviews use more AI credits, and may consume marginally more GitHub Actions minutes, than Lite reviews.\n\nUse Balanced for security-sensitive code, multi-service pull requests, or repositories with strict quality standards. Use Lite for routine changes where fast feedback is more important than exhaustive analysis.\n\nYou can select the review effort level when requesting a review in the pull request, under the **Reviewers** section where Copilot appears as a reviewer. Organization owners can set a default review effort level for automatic code reviews in their organization. Repository administrators can override the organization default for a specific repository.\n\nAfter Copilot code review reviews a pull request, the pull request overview comment shows the effort level used for each review run.\n\n## [Copilot approvals](https://docs.github.com/en/copilot/concepts/agents/code-review\\#copilot-approvals)\n\nNote\n\nCopilot approvals are in public preview and subject to change.\n\nEvery Copilot code review includes an approval assessment in the overview comment, indicating whether Copilot has determined the pull request ready to approve after reviewing it. By default, Copilot's reviews do not count toward required approvals for the pull request.\n\nWhen Copilot approvals are enabled in repository, organization, and enterprise settings, Copilot can submit an approving review that satisfies your repository's required-approval rule the same way a teammate's approval would. If new commits are pushed after Copilot approves, the approval is dismissed, and you can re-request a review.\n\n## [MCP servers and agent skills for code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#mcp-servers-and-agent-skills-for-code-review)\n\nCopilot code review can use repository-level agent skills and MCP servers when they are relevant to the review.\n\nCopilot code review is more likely to use skills and MCP context when your repository or pull request gives clear signals, including review-focused skill directory names, custom instructions that reference MCP context, and pull request descriptions that include identifiers referencing configured MCP servers such as issue keys or incident IDs.\n\n### [Agent skills](https://docs.github.com/en/copilot/concepts/agents/code-review\\#agent-skills)\n\nIf your repository includes agent skills, Copilot code review can automatically use relevant skills when reviewing a pull request, extending Copilot beyond its built-in analysis.\n\nWhen reviewing a pull request, Copilot reads repository custom instructions, agent instructions, and agent skills from the head branch (the branch with your changes), not the base branch. For example, when merging `my-feature-branch` into `main`, Copilot uses the instructions and skills in `my-feature-branch`, so you can test changes to them in the same pull request without merging them first.\n\nFor setup details, see [Adding agent skills for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/add-skills).\n\n### [MCP servers](https://docs.github.com/en/copilot/concepts/agents/code-review\\#mcp-servers)\n\nCopilot code review can use MCP servers to pull context directly into the review from the third-party platforms and internal systems your team uses, including issue tracking, documentation, service catalogs, and incident tooling.\n\nThe GitHub MCP server and Playwright MCP server are enabled by default.\n\nYou can configure MCP servers in your repository settings. Repository MCP configuration on GitHub applies to both Copilot cloud agent and Copilot code review. Changes you make to repository MCP settings affect both features. For setup details, see [Configure MCP servers for your repository](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers).\n\nIn repository settings, **Allow Copilot to use MCP tools when reviewing pull requests** is enabled by default. Disable this setting if you want MCP servers available only for Copilot cloud agent, and not for Copilot code review. For step-by-step instructions, see [Configure MCP servers for your repository](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers#disabling-mcp-tools-for-code-review).\n\n## [Excluded files](https://docs.github.com/en/copilot/concepts/agents/code-review\\#excluded-files)\n\nSome file types are excluded from Copilot code review:\n\n- Dependency management files, such as package.json and Gemfile.lock\n- Log files\n- SVG files\n\nIf you include these file types in a pull request, Copilot code review will not review the file.\n\nFor more information, see [Files excluded from GitHub Copilot code review](https://docs.github.com/en/copilot/reference/review-excluded-files).\n\n## [Enhancing Copilot's knowledge of a repository](https://docs.github.com/en/copilot/concepts/agents/code-review\\#enhancing-copilots-knowledge-of-a-repository)\n\nThe more Copilot knows about the code in your repository, the tools you use, and your coding standards and practices, the more accurate and useful its reviews will become. You can enhance Copilot's knowledge of your repositories in two ways.\n\n### [Custom instructions](https://docs.github.com/en/copilot/concepts/agents/code-review\\#custom-instructions)\n\nThese are short, natural-language statements that you write and store as one or more files in a repository. If you are the owner of an organization on GitHub, you can also define custom instructions in the settings for your organization. For more information, see [About customizing GitHub Copilot responses](https://docs.github.com/en/copilot/concepts/prompting/response-customization?tool=webui#about-repository-custom-instructions).\n\n### [Choosing between custom instructions, AGENTS.md, and skills](https://docs.github.com/en/copilot/concepts/agents/code-review\\#choosing-between-custom-instructions-agentsmd-and-skills)\n\nCopilot code review can draw on several sources of customization, and each serves a different purpose. Use `.github/copilot-instructions.md` for repository-wide rules specific to Copilot, use path-specific `*.instructions.md` files under `.github/instructions/` for rules that apply only to certain files or directories, use `AGENTS.md` for standing rules you want to share across AI tools and agents, and use skills for task-specific workflows that Copilot runs on demand. The following table summarizes when to use each and how to provide rules.\n\n| Use | `copilot-instructions.md` | Path-specific `*.instructions.md` | `AGENTS.md` | Skills |\n| --- | --- | --- | --- | --- |\n| Best for | Repository-wide, always-on rules for Copilot | Always-on rules for specific paths, file types, or directories | Always-on rules shared across AI agents | Task-specific review workflows |\n| Stored in | `.github/copilot-instructions.md` | `.github/instructions/**/*.instructions.md` | Repository root (`AGENTS.md`) | `.github/skills/...` |\n| Examples | Coding standards, architecture defaults, test expectations | `content/**` writing rules, `src/**` coding conventions, language-specific guidance | Shared repository conventions that should apply beyond Copilot | Reviews, releases, migrations, analysis |\n| Activation | Automatic | Automatic when the changed files match the instruction scope | Automatic (read from repository root) | Automatic when relevant (e.g. review-focused skills such as `code-review`), or on demand |\n| Scope | Repository-wide and Copilot-specific | Repository sub-paths and Copilot-specific | Cross-tool / agent-agnostic | Invoked per task |\n| Rule | \"Copilot, always know this for this repository\" | \"Copilot, always know this when working in these paths\" | \"Any agent, always know this\" | \"Do this when needed\" |\n\nFor more information, see [Adding agent skills for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/add-skills) and [Adding repository custom instructions for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/add-custom-instructions/add-repository-instructions).\n\n### [Copilot Memory (public preview)](https://docs.github.com/en/copilot/concepts/agents/code-review\\#copilot-memory-public-preview)\n\nIf you have a Copilot Pro, Copilot Pro+, or Copilot Max plan, you can enable Copilot Memory. This allows Copilot to store useful details it has learned about a repository. Copilot can then use this information when it reviews pull requests in that repository. For more information, see [About GitHub Copilot Memory](https://docs.github.com/en/copilot/concepts/agents/copilot-memory).\n\n## [Getting detailed code quality feedback across your repository](https://docs.github.com/en/copilot/concepts/agents/code-review\\#getting-detailed-code-quality-feedback-across-your-repository)\n\nGitHub Copilot code review reviews the changes in a pull request and suggests fixes. To add systematic feedback on the reliability and maintainability of your code, on pull requests and across your default branch, enable GitHub Code Quality.\n\nGitHub Code Quality complements Copilot code review by adding:\n\n- **Rules-based CodeQL-powered analysis** on pull requests and on your default branch.\n- **Test-coverage metrics** on pull requests, so you can see whether a change maintains or reduces coverage.\n- **One-click, Copilot-powered fixes**, including delegating remediation to Copilot cloud agent.\n- **Optional merge gating** with rulesets, so pull requests with unresolved rules-based findings (or that miss a coverage threshold) can be blocked from merging.\n\nFor more information, see [GitHub Code Quality](https://docs.github.com/en/code-security/concepts/code-quality/code-quality).\n\n## [Validating Copilot code reviews](https://docs.github.com/en/copilot/concepts/agents/code-review\\#validating-copilot-code-reviews)\n\nCopilot is not guaranteed to spot all problems or issues in a pull request. Sometimes it will make mistakes. Always validate Copilot's feedback carefully. Supplement Copilot's feedback with a human review.\n\nFor more information, see [Application card: GitHub Copilot Agents](https://docs.github.com/en/copilot/responsible-use/agents).\n\n## [Next steps](https://docs.github.com/en/copilot/concepts/agents/code-review\\#next-steps)\n\n- [Using GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review)\n- [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review)",
      "content_chars": 24904,
      "published_date": null
    },
    {
      "rank": 3,
      "url": "https://docs.github.com/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
      "title": "About GitHub Copilot cloud agent - GitHub Enterprise Cloud Docs",
      "content": "[Skip to main content](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent#main-content)\n\n[Skip to content](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article\n\n# About GitHub Copilot cloud agent\n\nCopilot can research a repository, create an implementation plan, and make code changes on a branch. You can review the diff, iterate, and create a pull request when you're ready.\n\nCopy markdown\n\n## Who can use this feature?\n\nCopilot cloud agent is available for all paid Copilot plans.\n\nThe agent is available in all repositories stored on GitHub, except repositories owned by managed user accounts and where it has been explicitly disabled.\n\n[Sign up for Copilot](https://github.com/features/copilot/plans?ref_product=copilot&ref_type=engagement&ref_style=button)\n\n## In this article\n\n## [Overview of Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#overview-of-copilot-cloud-agent)\n\nWith Copilot cloud agent, GitHub Copilot can work independently in the background to complete tasks, just like a human developer.\n\nCopilot cloud agent can:\n\n- Research a repository\n- Create implementation plans\n- Fix bugs\n- Implement incremental new features\n- Improve test coverage\n- Update documentation\n- Address technical debt\n- Resolve merge conflicts\n\nWhen you delegate tasks to Copilot cloud agent, you can:\n\n- Use the agents panel or other agents entry points on GitHub.com to have Copilot research, plan, and make code changes on a branch, then iterate before creating a pull request. You can also specify in your prompt that you want a pull request created right away. See [Research, plan, and iterate on code changes with Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/use-copilot-agents/research-plan-iterate).\n- Ask Copilot to open a new pull request from other entry points, including GitHub Issues and Visual Studio Code. See [Starting GitHub Copilot sessions](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent/start-copilot-sessions).\n- Mention `@copilot` in a comment on an existing pull request to ask it to make changes. See [Using Copilot cloud agent on GitHub](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent/use-cloud-agent-on-github).\n- Set up an automation to run Copilot automatically, on a schedule or in response to events such as an issue being opened. See [About Copilot automations](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-automations).\n- Assign security alerts to Copilot from security campaigns. See [Fixing alerts in a security campaign](https://docs.github.com/en/enterprise-cloud@latest/code-security/how-tos/manage-security-alerts/remediate-alerts-at-scale/fixing-alerts-in-security-campaign#assigning-alerts-to-copilot-cloud-agent).\n\nCopilot cloud agent will evaluate the task it has been assigned based on the prompt you give it.\n\nWhile working on a coding task, Copilot cloud agent has access to its own ephemeral development environment, powered by GitHub Actions, where it can explore your code, make changes, execute automated tests and linters and more.\n\nNote\n\nDeep research, planning, and iterating on code changes before creating a pull request are only available with Copilot cloud agent on GitHub.com, and in public preview for the Microsoft Teams and Slack integrations. Other cloud agent integrations (such as Azure Boards, JIRA, or Linear) only support creating a pull request directly.\n\n### [Benefits over traditional AI workflows](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#benefits-over-traditional-ai-workflows)\n\nWhen used effectively, Copilot cloud agent offers productivity benefits over traditional AI assistants in IDEs:\n\n- With **AI assistants in IDEs**, coding happens **locally**. Individual developers pair in **synchronous** sessions with the AI assistant. Decisions made during the session are **untracked** and lost to time unless committed. Although the assistant helps write code, the developer still has a lot of **manual steps** to do: create the branch, write commit messages, push the changes, open the PR, write the PR description, get a review, iterate in the IDE, and repeat. These steps take time and effort that may be hard to justify for simple or routine issues.\n\n- With **Copilot cloud agent**, all coding and iterating happens **on GitHub**. You can ask Copilot to **research** a repository, **create a plan**, and **make code changes** on a branch\u2014all before opening a pull request. You can create multiple custom agents that specialize in different types of tasks. Copilot **automates** branch creation, commit message writing, and pushing. Developers let the agents **work in the background** and then chooses to **create a pull request** when ready. Working on GitHub adds **transparency**, with every step happening in a commit and being viewable in logs, and opens up **collaboration** opportunities for the entire team.\n\n\n### [Sharing context between Copilot Chat and Copilot cloud agent sessions](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#sharing-context-between-copilot-chat-and-copilot-cloud-agent-sessions)\n\nWhen you start a Copilot cloud agent session from Copilot Chat on GitHub, the session carries the context of your chat. You can continue chatting while the session runs, and ask follow-up questions about progress in the same conversation. For more information, see [Managing agent sessions](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/use-copilot-agents/manage-and-track-agents).\n\n## [Copilot cloud agent versus agent mode](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#copilot-cloud-agent-versus-agent-mode)\n\nCopilot cloud agent is distinct from the \"agent mode\" feature available in your IDE. Copilot cloud agent works autonomously in a GitHub Actions-powered environment to complete development tasks assigned through GitHub issues or GitHub Copilot Chat prompts. It can research a repository, create a plan, make code changes on a branch, and optionally open a pull request. In contrast, agent mode in your IDE makes autonomous edits directly in your local development environment. For more information about agent mode, see [Asking GitHub Copilot questions in your IDE](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/chat-with-copilot/chat-in-ide).\n\n## [Streamlining software development with Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#streamlining-software-development-with-copilot-cloud-agent)\n\nAssigning tasks to Copilot cloud agent can enhance your software development workflow.\n\nFor example, you can assign Copilot cloud agent to straightforward issues on your backlog by selecting \"Copilot\" as the assignee. This allows you to spend less time on these issues and more time on more complex or interesting work, or work that requires a high degree of creative thinking. Copilot cloud agent can work on \"nice to have\" issues that improve the quality of your codebase or product, but often remain on the backlog while you focus on more urgent work.\n\nHaving Copilot cloud agent as an additional coding resource also allows you to start tasks that you might not have otherwise started due to lack of resources. For example, you might create issues to refactor code or add more logging, and then immediately assign these to Copilot.\n\nYou can also use Copilot cloud agent to research a repository and create a plan before any code is written, helping you understand how a codebase works or agree on an approach before committing to changes. See [Research, plan, and iterate on code changes with Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/use-copilot-agents/research-plan-iterate).\n\nCopilot cloud agent can start a task, which you then pick up and continue working on yourself. By assigning the initial work to Copilot, you free up time that you would otherwise have spent doing repetitive tasks, such as setting up the scaffolding for a new project.\n\nYou can create specialized custom agents for different tasks. For example, you might create a custom agent specialized for frontend development that focuses on React components and styling, a documentation agent that excels at writing and updating technical documentation, or a testing agent that specializes in generating comprehensive unit tests. Each custom agent can be tailored with specific prompts and tools suited to its particular task.\n\n## [Measuring pull request outcomes for Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#measuring-pull-request-outcomes-for-copilot-cloud-agent)\n\nEnterprise administrators and organization owners can use Copilot usage metrics to analyze pull request outcomes for pull requests created by Copilot cloud agent.\n\nThe Copilot usage metrics APIs include pull request lifecycle metrics such as:\n\n- The total number of pull requests created and merged\n- The number of pull requests created by Copilot cloud agent that have been merged\n- Median time to merge for merged pull requests, including pull requests created by Copilot cloud agent\n\nThese metrics can help you track adoption of Copilot cloud agent and monitor changes in pull request throughput and time to merge over time. See [GitHub Copilot usage metrics](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/copilot-usage-metrics/copilot-metrics).\n\n## [Integrating Copilot cloud agent with third-party tools](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#integrating-copilot-cloud-agent-with-third-party-tools)\n\nYou can invoke Copilot cloud agent from external tools, allowing you to assign tasks to Copilot, provide context, and open pull requests without leaving your workflow.\n\nUse Copilot cloud agent in Microsoft Teams and Slack to collaborate with your team on agent-assisted work. You can @mention GitHub in channels, threads, and direct messages to work alongside teammates and Copilot on research, planning, and coding tasks. Teammates can add context, steer Copilot sessions, monitor progress, and then review the resulting artifacts.\n\nFor more information, see [About Copilot integrations](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/tools/about-copilot-integrations).\n\n## [Making Copilot cloud agent available](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#making-copilot-cloud-agent-available)\n\nBefore you can assign tasks to Copilot cloud agent, it must be enabled. Copilot cloud agent is available for all paid Copilot plans.\n\nIf you are a GitHub Copilot Business or GitHub Copilot Enterprise subscriber, an administrator must enable the relevant policy before you can use the agent.\n\nRepository owners can choose to opt out some or all repositories from Copilot cloud agent.\n\nSee [Managing access to GitHub Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/enterprise/cloud-agent-access).\n\n## [AI models for Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#ai-models-for-copilot-cloud-agent)\n\nDepending on how you start your Copilot cloud agent task, you may be able to select the model used by Copilot cloud agent. You may find that different models perform better, or provide more useful responses, depending on the type of tasks you give Copilot.\n\nSee [Changing the AI model for GitHub Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent/changing-the-ai-model).\n\n## [Enhancing Copilot cloud agent's knowledge of a repository](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#enhancing-copilot-cloud-agents-knowledge-of-a-repository)\n\nThe more Copilot cloud agent knows about the code in your repository, the tools you use, and your coding standards and practices, the more effective it will become. There are two ways you can enhance Copilot cloud agent's knowledge of a repository.\n\n- **Custom instructions**\n\nThese are short, natural\u2011language statements that you write and store as one or more files in a repository. If you are the owner of an organization on GitHub you can also define custom instructions in the settings for your organization. See [About customizing GitHub Copilot responses](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/prompting/response-customization?tool=webui#about-repository-custom-instructions).\n\n- **Copilot Memory** (public preview)\n\nIf you have a Copilot Pro, Copilot Pro+, or Copilot Max plan, you can enable Copilot Memory. This allows Copilot to store useful details it has worked out for itself about a repository. Copilot cloud agent can then use this information when it is working in that repository. See [About GitHub Copilot Memory](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/copilot-memory).\n\n\n## [Copilot cloud agent usage costs](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#copilot-cloud-agent-usage-costs)\n\nCopilot cloud agent uses GitHub Actions minutes and AI credits. The AI credits consumed depend on the model used and the number of tokens processed during the session.\n\nWithin your included GitHub Actions minutes and AI credits, you can use Copilot cloud agent without incurring additional costs. See [Usage-based billing for organizations and enterprises](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/billing/organizations-and-enterprises/usage-based-billing).\n\nCopilot code review also consumes GitHub Actions minutes on private repositories. See [GitHub Actions billing](https://docs.github.com/en/enterprise-cloud@latest/billing/concepts/product-billing/github-actions#copilot-code-review-and-github-actions-minutes).\n\n## [Customizing Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#customizing-copilot-cloud-agent)\n\nYou can customize Copilot cloud agent in a number of ways:\n\n- **Custom instructions**: Custom instructions allow you to give Copilot additional context on your project and how to build, test and validate its changes. See [Adding repository custom instructions for GitHub Copilot](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/add-custom-instructions/add-repository-instructions).\n- **Model Context Protocol (MCP) servers**: MCP servers allow you to give Copilot access to different data sources and tools. Repository MCP settings on GitHub apply to both Copilot cloud agent and Copilot code review. The GitHub MCP server and Playwright MCP server are enabled by default for both features. See [Configure MCP servers for your repository](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers).\n- **Custom agents**: Custom agents allow you to create different specialized versions of Copilot for different tasks. For example, you could customize Copilot to be an expert frontend engineer following your team's guidelines. See [About custom agents](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-custom-agents).\n- **Hooks**: Hooks allow you to execute custom shell commands at key points during agent execution, enabling you to add validation, logging, security scanning, or workflow automation. See [About hooks for GitHub Copilot](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/hooks).\n- **Skills**: Skills allow you to enhance the ability of Copilot to perform specialized tasks with instructions, scripts, and resources. See [About agent skills](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/about-agent-skills).\n\n## [Limitations of Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#limitations-of-copilot-cloud-agent)\n\nCopilot cloud agent has certain limitations in its software development workflow and compatibility with other features.\n\n### [Limitations in Copilot cloud agent's software development workflow](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#limitations-in-copilot-cloud-agents-software-development-workflow)\n\n- **Copilot can only make changes in the repository specified when you start a task**. Copilot cannot make changes across multiple repositories in one run.\n- **By default, Copilot can only access context in the repository specified when you start a task**. The Copilot MCP server is configured by default to allow Copilot to access context (for example issues and historic pull requests) in the repository where it is working. You can, however, configure broader access through repository MCP settings. See [Configure MCP servers for your repository](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers).\n- **Copilot can only work on one branch at a time** and can open exactly one pull request to address each task it is assigned.\n- **Each Copilot cloud agent session has a maximum execution time of 59 minutes**. This is a hard limit that cannot be extended or bypassed. If a task exceeds this limit, the session will time out and stop. For complex tasks that may require more time, consider breaking the work into smaller, more focused tasks. You can configure a shorter timeout using the `timeout-minutes` setting in your `copilot-setup-steps.yml` file. See [Configure the development environment](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/customize-the-agent-environment).\n\n### [Limitations in Copilot cloud agent's compatibility with other features](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#limitations-in-copilot-cloud-agents-compatibility-with-other-features)\n\n- **Copilot isn't able to comply with certain rules that may be configured for your repository**. If you have configured a ruleset or branch protection rule that isn't compatible with Copilot cloud agent, access to the agent will be blocked. For example, a rule that only allows specific commit authors can prevent Copilot cloud agent from creating or updating pull requests. If the rule is configured using rulesets, you can add Copilot as a bypass actor to enable access. See [Creating rulesets for a repository](https://docs.github.com/en/enterprise-cloud@latest/repositories/configuring-branches-and-merges-in-your-repository/managing-rulesets/creating-rulesets-for-a-repository#granting-bypass-permissions-for-your-branch-or-tag-ruleset).\n- **Copilot cloud agent only works with repositories hosted on GitHub**. If your repository is stored using a different code hosting platform, Copilot won't be able to work on it.\n\n## [Hands-on practice](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#hands-on-practice)\n\nTry the [Expand your team with Copilot cloud agent](https://github.com/skills/expand-your-team-with-copilot/?ref_product=copilot&ref_type=engagement&ref_style=text) Skills exercise for practical experience with Copilot cloud agent.\n\n## [Further reading](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#further-reading)\n\n- [GitHub Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent) how-to articles\n- [About custom agents](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-custom-agents)\n- [Application card: GitHub Copilot Agents](https://docs.github.com/en/enterprise-cloud@latest/copilot/responsible-use/agents)",
      "content_chars": 20773,
      "published_date": null
    },
    {
      "rank": 4,
      "url": "https://docs.github.com/en/copilot/concepts/learning-about-new-features-and-models",
      "title": "Learning about new features and models - GitHub Docs",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article\n\n# Learning about new features and models\n\nStay informed about GitHub Copilot features and models, and make confident decisions about enabling them for your enterprise.\n\nCopy markdown\n\n## In this article\n\n## Up next\n\n[Pilot a new Copilot feature or model in your enterprise](https://docs.github.com/en/copilot/tutorials/roll-out-at-scale/govern-at-scale/pilot-a-feature-or-model)\n\nGitHub Copilot is evolving rapidly, with new features and models released regularly. If you're an enterprise administrator, staying informed helps you make confident decisions about which capabilities to enable, when to adopt them, and how to manage risk across your organizations.\n\nThis article provides guidance on navigating the Copilot landscape, understanding different types of features and models, and preparing for upcoming changes.\n\n## [Keeping up to date with Copilot changes](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#keeping-up-to-date-with-copilot-changes)\n\nYou can use GitHub Docs and the GitHub Blog to keep track of new releases.\n\n### [Learning about new Copilot features](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#learning-about-new-copilot-features)\n\nTo learn about new Copilot features, we recommend monitoring these two locations:\n\n- **Copilot blog updates**: For broader news and updates about Copilot, see the [Copilot blog](https://github.blog/ai-and-ml/github-copilot/).\n- **Changelog**: For announcements about newly released and updated features, follow the [Copilot changelog](https://github.blog/changelog/?label=copilot).\n\nCopilot's capabilities and supporting features fall into four categories:\n\n| Type | Description | Examples | Considerations |\n| --- | --- | --- | --- |\n| Assistive | Respond to prompts and provide suggestions, but require human review before changes are made. | Inline suggestions, Copilot Chat, Copilot pull request summaries. | All changes require user approval. |\n| Agentic | Autonomously research, plan, and make changes on behalf of users. | Copilot cloud agent, third-party agents. | Can work autonomously, but with built-in protections. |\n| Customizations | Add context, instructions, tools, skills, and agents to tailor how Copilot responds and works. | Custom instructions, prompt files, Copilot Spaces, MCP servers, agent skills, custom agents. | Support and configuration vary by feature and surface. Organization and enterprise policies can control access to some customizations. |\n| External AI agents, models, and tools | External coding agents, models, and tools, including MCP servers from other providers, that work alongside Copilot cloud agent to complete tasks asynchronously. | Anthropic Claude, OpenAI Codex. | Same as agentic. Review provider documentation and enable via policies. See [About third-party coding agents](https://docs.github.com/en/copilot/concepts/agents/about-third-party-coding-agents). |\n\nEach feature has its own enablement requirements and policy settings. When a new feature is released:\n\n1. Review the feature documentation to understand its capabilities.\n2. Check the policy settings available at the enterprise and organization level. See [GitHub Copilot policies for enterprises and organizations](https://docs.github.com/en/copilot/concepts/enterprise/policies).\n3. Consider running a pilot with a subset of users before broader rollout.\n\n### [Learning about new Copilot models](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#learning-about-new-copilot-models)\n\nGitHub Copilot uses multiple AI models from different providers to power its features. Different models have different strengths: some prioritize speed and cost-efficiency, while others are optimized for accuracy, reasoning, or working with multimodal inputs (like images and code together).\n\nUsers will have access to different models depending on the feature they're using and your enterprise's Copilot plan and policies.\n\nYou can find information about the models available and upcoming models in the following locations:\n\n- **Supported models**: For a complete list of available models and their capabilities, see [Supported AI models in GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/supported-models).\n- **Model comparison**: To compare model capabilities side by side, see [AI model comparison](https://docs.github.com/en/copilot/reference/ai-models/model-comparison).\n- **Changelog**: Model updates are announced in the [Copilot changelog](https://github.blog/changelog/label/copilot/).\n\n#### [Default enablement](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#default-enablement)\n\nFor enterprises on a Copilot Business or Copilot Enterprise plan, a policy controls whether unconfigured generally available (GA) models are enabled or disabled by default. See [About default availability of Copilot models](https://docs.github.com/en/copilot/concepts/enterprise/default-model-availability).\n\n#### [Special categories](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#special-categories)\n\nGitHub categorizes certain types of model, allowing you to plan for model transitions and set user expectations.\n\n| Model type | Description | Why it matters |\n| --- | --- | --- |\n| Utility model | A small set of models that power background GitHub features across surfaces. | Your enterprise cannot disable these models. Monitor these to ensure they are compliant. |\n| Base model | The default model when no other models are enabled. | Automatically enabled within 60 days of designation. |\n| LTS model | A model supported for one year from designation. | Allows enterprises to build workflows around a stable model. |\n\nFor more information, see [Utility models](https://docs.github.com/en/copilot/concepts/models/utility-models) and [Base and long-term support (LTS) models](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models).\n\n## [Considering different release stages](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#considering-different-release-stages)\n\nCopilot features and models progress through different release stages. Understanding release stages helps you decide when to enable features for your organization. Each stage has different expectations for stability and support.\n\n| Stage | Availability | Stability |\n| --- | --- | --- |\n| Private preview | Limited set of customers by invitation. | May change significantly based on feedback. |\n| Public preview | All eligible customers who opt in. | Functional but may have limitations or change. |\n| Generally available (GA) | All customers. | Fully supported and recommended for production. |\n\nTip\n\nMany preview features are covered by the [GitHub Data Protection Agreement](https://gh.io/dpa), meaning you can test them without breaching compliance requirements. See [GitHub DPA-Covered Previews](https://gh.io/dpa-previews) for a list.\n\nPreview features are controlled by policy settings at the organization and enterprise level. The **Opt in to preview features** policy allows administrators to enable or disable preview features on the GitHub website specifically. See [Managing policies and features for GitHub Copilot in your enterprise](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-enterprise/manage-enterprise-policies) and [Managing policies and features for GitHub Copilot in your organization](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-organization/manage-policies).\n\n## [Reviewing information about Responsible AI](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#reviewing-information-about-responsible-ai)\n\nGitHub is committed to responsible AI development and provides transparency documentation for each Copilot feature.\n\nEach Copilot feature has a dedicated **responsible use** article describing its purpose, capabilities, and limitations. See [Responsible use of GitHub Copilot features](https://docs.github.com/en/copilot/responsible-use).",
      "content_chars": 8569,
      "published_date": null
    },
    {
      "rank": 5,
      "url": "https://docs.github.com/en/copilot/responsible-use/agents",
      "title": "Application card: GitHub Copilot Agents",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/responsible-use/agents#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/responsible-use/agents#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article8\\. Evaluations\n\n# Application card: GitHub Copilot Agents\n\nLearn how to use GitHub Copilot agentic features responsibly by understanding their purposes, capabilities, and limitations.\n\nCopy markdown\n\n## In this article\n\n## [What is an Application Card?](https://docs.github.com/en/copilot/responsible-use/agents\\#what-is-an-application-card)\n\nGitHub\u2019s application and platform cards are intended to help you understand how our AI technology works, the choices application owners can make that influence application performance and behavior, and the importance of considering the whole application, including the technology, the people, and the environment. Application cards are created for AI applications and platform cards are created for AI platform services. These resources can support the development or deployment of your own applications and can be shared with users or stakeholders impacted by them.\n\nAs part of its commitment to responsible AI, GitHub adheres to Microsoft's [six core principles](https://www.microsoft.com/en-us/ai/principles-and-approach/?msockid=3da790040c776d6f2b5485e40de56c06#ai-principles): fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability. These principles are embedded in the [Responsible AI Standard](https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/final/en-us/microsoft-brand/documents/Microsoft-Responsible-AI-Standard-General-Requirements.pdf?culture=en-us&country=us), which guides teams in designing, building, and testing AI applications. Application and Platform Cards play a key role in operationalizing these principles by offering transparency around capabilities, intended uses, and limitations. For further insight, readers are encouraged to explore Microsoft\u2019s [Responsible AI Transparency Report](https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Responsible-AI-Transparency-Report-2025-vertical.pdf) and [GitHub Terms](https://docs.github.com/en/site-policy/github-terms).\n\n## [1\\. Overview](https://docs.github.com/en/copilot/responsible-use/agents\\#1-overview)\n\nGitHub Copilot includes several agentic features that go beyond suggestion and conversation\u2014they can review code, take action on your behalf, and build applications. This card covers the following experiences:\n\n- **Copilot code review**: Reviews pull request diffs and metadata on GitHub.com, producing feedback comments and suggested changes.\n- **Copilot cloud agent**: An asynchronous agent on GitHub.com that can create branches, write code, and open pull requests in response to assigned issues. The cloud agent runs in an ephemeral, firewalled environment with automated security scanning.\n- **Copilot CLI**: A command-line tool that can create and modify files, execute commands, and perform multi-step tasks. All actions require explicit permission prompts and are scoped to the current directory.\n- **Copilot SDK**: A programmatic library that allows developers to build custom AI-powered applications using Copilot. The SDK communicates with Copilot CLI over JSON-RPC and supports custom agents, MCP server integrations, lifecycle hooks, and session management.\n- **GitHub Copilot app**: A desktop application for directing agent sessions across local repositories, git worktrees, and cloud sandboxes. The app uses GitHub Copilot CLI and GitHub Copilot SDK as its foundation and adds a user interface for parallel sessions, quick chats, GitHub issue and pull request workflows, automations, and canvases.\n\nThese features share common principles\u2014human oversight, review of outputs, and responsible use\u2014but differ in their execution environments, permissions, and data flows. The sections below describe each experience in context.\n\n## [2\\. Key terms](https://docs.github.com/en/copilot/responsible-use/agents\\#2-key-terms)\n\nThe following list provides a glossary of key terms related to GitHub Copilot Agents:\n\n- **Code suggestion**: A specific code change proposed by Copilot code review as part of its feedback on a pull request. Code suggestions are presented as suggested changes that can be applied with a couple of clicks.\n- **Content filtering**: A safety system that scans prompts and responses to detect and block harmful, offensive, or insecure content before it is shown to the user.\n- **Custom instructions**: Natural language descriptions of coding style and best practices that a repository maintainer can configure to guide Copilot code review's feedback. Custom instructions help Copilot understand the conventions and standards of a specific codebase.\n- **Hallucination**: A phenomenon where a language model generates output that sounds plausible but is factually incorrect, unsupported by the provided context, or entirely fabricated. In code review, hallucination can manifest as feedback that highlights problems that do not exist or are based on misunderstandings of the code.\n- **Large language model (LLM)**: A type of neural network trained on a large body of text data that can generate, analyze, and transform natural language and code. Copilot Agents use one or more LLMs to process context and produce responses.\n- **Ephemeral development environment**: A temporary, isolated compute environment created for each cloud agent session. The environment is destroyed after the session ends, ensuring that no state persists between runs.\n- **Firewall**: A network-level control enabled by default for the cloud agent that prevents outbound connections to unauthorized hosts, protecting against accidental or malicious exfiltration of code or data.\n- **AI credit**: A unit of consumption for Copilot features. Each use of Copilot code review consumes AI credits.\n- **Permission prompt**: An interactive confirmation step in Copilot CLI that asks the user to approve an action\u2014such as modifying a file, executing a command, or accessing files outside the current directory\u2014before the agent proceeds. Permission prompts are a key safety mechanism for local agentic execution.\n- **Copilot SDK**: A programmatic library (`@github/copilot-sdk`) that lets developers build custom AI-powered applications by creating sessions, sending prompts, and receiving streaming responses from Copilot. The SDK communicates with Copilot CLI over JSON-RPC.\n- **JSON-RPC**: The communication protocol used between the Copilot SDK and Copilot CLI. The SDK sends structured requests to the CLI process, which handles model interaction and tool execution.\n- **Custom agent (SDK)**: A named agent configuration within the SDK that has its own system prompt, scoped tools, and optional MCP servers. The SDK runtime can automatically delegate to sub-agents based on user intent.\n- **Hooks (SDK)**: Lifecycle callbacks in the Copilot SDK that let developers inject custom logic at specific points during a session, such as before or after tool use, on session start or end, and on error.\n\n## [3\\. Key features or capabilities](https://docs.github.com/en/copilot/responsible-use/agents\\#3-key-features-or-capabilities)\n\nThe key features and capabilities outlined here describe what GitHub Copilot Agents are designed to do and how they perform across supported tasks.\n\n- **Automated code review feedback**: When a user requests a review from Copilot, it scans the code changes plus additional relevant context and provides feedback on the code. Feedback may include natural language comments and specific code suggestions linked to particular lines and files.\n- **Customizable review guidance**: Copilot code review can be customized with custom instructions\u2014natural language descriptions of coding style and best practices\u2014so that feedback reflects a repository's conventions and standards.\n- **Autonomous pull request creation**: The Copilot cloud agent can pick up a task from an issue, a pull request comment, or Copilot Chat, create a branch, generate tailored code changes, and open a pull request. After the initial pull request is created, the agent can iterate based on your feedback and reviews.\n- **Ephemeral, firewalled execution**: While working on a task, the cloud agent has access to its own ephemeral development environment where it can make changes to code, execute automated tests, and run linters. A firewall is enabled by default to prevent data exfiltration.\n- **Automated security scanning**: During code generation, the cloud agent automatically analyzes newly generated code for security vulnerabilities using CodeQL, secret scanning, and dependency analysis, and attempts to resolve any issues before they are introduced.\n- **External integrations**: The cloud agent can receive information and context from MCP like workIQ and Microsoft 365, and external applications like Microsoft Teams, Linear, Slack, and Jira, enabling teams to assign tasks and track progress directly within their existing workflows.\n- **Local agentic execution (Copilot CLI)**: Copilot CLI provides a chat-like interface in the terminal that can autonomously create and modify files, execute commands, and perform multi-step tasks. All actions are scoped to the current directory and require explicit permission prompts before the agent modifies files or runs commands.\n\n## [4\\. Intended uses](https://docs.github.com/en/copilot/responsible-use/agents\\#4-intended-uses)\n\nGitHub Copilot Agents can be used in multiple scenarios across a variety of industries. Some examples of use cases include:\n\n- **Supplementing human code review**: Copilot code review is intended to quickly provide feedback on a developer's code, enabling developers to get code ready to merge more quickly and increasing overall code quality.\n- **Codebase maintenance**: The cloud agent can tackle security-related fixes, dependency upgrades, and targeted refactoring.\n- **Feature development**: The cloud agent can implement incremental feature requests, develop additional test suites, and create or update documentation.\n- **Prototyping new projects**: The cloud agent and Copilot CLI can greenfield new concepts, helping developers explore ideas quickly.\n- **Setting up your environment (CLI)**: Copilot CLI can run commands in your terminal to set up your local environment to work on existing projects.\n- **Finding the right command (CLI)**: Copilot CLI can suggest commands to perform tasks you're trying to complete, and explain unfamiliar commands in natural language.\n- **Building custom AI applications (SDK)**: The Copilot SDK enables developers to build applications that leverage Copilot for code generation, natural language interaction, and task automation in their own products and workflows.\n- **Multi-agent orchestration (SDK)**: Using custom agents and sub-agents, developers can build sophisticated workflows where multiple specialized agents collaborate on complex tasks, with automatic delegation based on user intent.\n- **Extending applications with external tools (SDK)**: The SDK's MCP server support allows developers to connect their applications to external data sources and services, expanding the range of tasks their agents can perform.\n\n## [5\\. Models and training data](https://docs.github.com/en/copilot/responsible-use/agents\\#5-models-and-training-data)\n\nGitHub Copilot Agents leverage a variety of AI models to power the experience that users see. For a comparison of the models available for Copilot, see [AI model comparison](https://docs.github.com/en/copilot/reference/ai-models/model-comparison). For the full list of supported models, see [Supported AI models in GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/supported-models). For information on where models are hosted, see [Hosting of models for GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/model-hosting). To learn more about the data used to train the foundation models behind GitHub Copilot Agents, refer to the linked AI model comparison above and [What data has GitHub Copilot been trained on?](https://github.com/features/copilot#faq) in the GitHub Copilot FAQ.\n\nCopilot code review is a purpose-built product that uses a carefully tuned mix of models, prompts, and system behaviors to deliver consistent, high-quality feedback across a wide range of codebases. Model switching is not supported, as changing the model is likely to compromise reliability, user experience, and the quality of review comments. Copilot code review may use models that are not enabled on your organization's \"Models\" settings page.\n\nThe Copilot cloud agent uses a large language model to reason about tasks, generate code, and leverage tools within its ephemeral development environment. The agent has been evaluated across a variety of programming languages. English is the primary supported language for prompts and responses.\n\nCopilot CLI uses a large language model to reason about tasks, generate code, modify files, and execute commands in your local terminal environment. The agent has been evaluated across a variety of programming languages. English is the primary supported language for prompts and responses.\n\nThe Copilot SDK communicates with Copilot CLI over JSON-RPC, using the same underlying models and capabilities. Applications built with the SDK use the same models available to the authenticated Copilot user or organization. Developers can also bring their own API keys (BYOK) to use custom model providers.\n\n## [6\\. Performance](https://docs.github.com/en/copilot/responsible-use/agents\\#6-performance)\n\n### [Differences by experience](https://docs.github.com/en/copilot/responsible-use/agents\\#differences-by-experience)\n\n#### [Copilot cloud agent](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cloud-agent)\n\nThe Copilot cloud agent works by using a combination of natural language processing and machine learning to understand your task and make changes in a codebase. This process can be broken down into a number of steps:\n\n1. **Prompt processing**: The task provided through an issue, pull request comment, or Copilot Chat message is combined with other relevant, contextual information to form a prompt. That prompt is sent to a large language model for processing. Inputs can take the form of plain natural language, code snippets, or images.\n2. **Language model analysis**: The prompt is passed through a large language model, which is a neural network that has been trained on a large body of data. The language model analyzes the input prompt to help the agent reason about the task and leverage necessary tools.\n3. **Response generation**: The language model generates a response based on its analysis of the prompt. This response can take the form of natural language suggestions and code suggestions.\n4. **Output formatting**: Once the agent completes its first run, it updates the pull request description with the changes it made. The agent may include supplemental information about resources it could not access and provide suggestions on steps to resolve. You may provide feedback by commenting within the pull request or explicitly mentioning the agent (`@copilot`). The agent will then resubmit that feedback for further analysis and respond with updated changes.\n\nCopilot cloud agent is intended to provide the most relevant solution for task resolution. However, it may not always provide the answer you are looking for. You are responsible for reviewing and validating responses generated by Copilot cloud agent to ensure they are accurate and appropriate.\n\n#### [Copilot code review](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-code-review)\n\nCopilot code review inspects your code and provides feedback using a combination of natural language processing and machine learning. This process can be broken down into a number of steps:\n\n1. **Input processing**: The code changes are combined with other relevant, contextual information (for example, the pull request's title and body), and any custom instructions that have been defined, to form a prompt. That prompt is sent to a large language model.\n2. **Language model analysis**: The prompt is passed through the Copilot language model, which is a neural network that has been trained on a large body of text data. The language model analyzes the input prompt.\n3. **Response generation**: The language model generates a response based on its analysis of the input prompt. This response can take the form of natural language suggestions and code suggestions.\n4. **Output formatting**: The response is presented to the user either directly in a supported editor or as a pull request review on GitHub.com, providing code feedback linked to specific lines of specific files. Where Copilot has provided a code suggestion, the suggestion is presented as a suggested change, which can be applied with a couple of clicks.\n\n#### [Copilot CLI](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cli)\n\nCopilot CLI works by using a combination of natural language processing and machine learning to understand your task and make changes in a codebase. This process can be broken down into a number of steps:\n\n1. **Input processing**: Your input is combined with relevant contextual information to form a prompt. That prompt is sent to a large language model for processing. Inputs can take the form of plain natural language, code snippets, or references to files in your terminal.\n2. **Language model analysis**: The prompt is passed through a large language model, which is a neural network that has been trained on a large body of data. The language model analyzes the input prompt to help the agent reason about the task and use the necessary tools.\n3. **Response generation**: The language model generates a response based on its analysis of the prompt. This response can take the form of natural language suggestions, code suggestions, file modifications, and command executions.\n4. **Output formatting**: The response is formatted and presented using syntax highlighting, indentation, and other formatting features. The agent may also execute commands in your local environment and create, edit, or delete files in your file system in order to complete the task. All such actions require explicit permission prompts.\n\nYou may provide feedback to the agent after it returns a response in the interactive chat window. The agent will resubmit that feedback to the language model for further analysis and return an additional response.\n\n#### [Copilot SDK](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-sdk)\n\nThe Copilot SDK provides a programmatic interface to Copilot's agentic capabilities. Applications built with the SDK follow this process:\n\n1. **Session creation**: The application creates a session with the SDK, specifying the model, system prompt, available tools, custom agents, MCP servers, and hooks. The SDK establishes a JSON-RPC connection to Copilot CLI.\n2. **Prompt submission**: User input is submitted to the session. The SDK routes the prompt (along with session context) to the CLI, which forwards it to the language model.\n3. **Agent execution**: The language model reasons about the task and may invoke tools, delegate to sub-agents, or connect to MCP servers. Lifecycle hooks fire at each stage, allowing the application to inject custom logic.\n4. **Response streaming**: Responses are streamed back to the application, which can present them in any format appropriate for its interface. The SDK provides structured events for text, tool calls, errors, and completion signals.\n\n## [7\\. Limitations](https://docs.github.com/en/copilot/responsible-use/agents\\#7-limitations)\n\nUnderstanding GitHub Copilot agentic features' limitations is crucial to determine they are used within safe and effective boundaries. While we encourage customers to leverage these features in their innovative solutions or applications, it's important to note that they were not designed for every possible scenario. We encourage users to refer to [GitHub Terms](https://docs.github.com/en/site-policy/github-terms) as well as the following considerations when choosing a use case:\n\n- **Missed code quality problems**: Copilot may not identify all of the problems that are present in code, especially where changes are large or complex. To ensure that all relevant problems are identified and corrected, Copilot code review should be supplemented with careful human code review.\n- **False positives**: Copilot code review has a risk of hallucination\u2014it may highlight problems in reviewed code that do not exist or are based on misunderstandings of the code. Comments generated by Copilot code review should be carefully reviewed and considered before taking action and making changes.\n- **Inaccurate or insecure code suggestions**: As part of its comments, Copilot code review may provide specific code suggestions. The code generated may appear to be valid but may not actually be semantically or syntactically correct, or may not correctly resolve the problem identified in the comment. In addition, code generated by Copilot may contain security vulnerabilities or other issues. You should always carefully review and test code generated by Copilot.\n- **Potential biases**: Copilot's training data is drawn from existing code repositories, which may contain biases and errors that can be perpetuated by the tool. Additionally, Copilot code review may be biased toward certain programming languages or coding styles, which can lead to suboptimal or incomplete feedback.\n- **Limited scope (cloud agent)**: The language model used by the cloud agent has been trained on a large body of code but still has a limited scope and may not be able to handle certain code structures or obscure programming languages. For each language, the quality of suggestions depends on the volume and diversity of training data for that language.\n- **Inaccurate code (cloud agent)**: The cloud agent may generate code that appears to be valid but may not actually be semantically or syntactically correct, or may not accurately reflect the intent of the developer. You should carefully review and test generated code, particularly when dealing with critical or sensitive applications.\n- **Security risks (cloud agent)**: The cloud agent generates code and natural language based on the context of an issue or comment within a repository, which can potentially expose sensitive information or vulnerabilities if not used carefully. You should review all outputs generated by the agent thoroughly prior to merging.\n- **Public code matches (cloud agent)**: The cloud agent may generate code that is a match or near match of publicly available code, even if the \"Suggestions matching public code\" policy is set to \"Block.\" If this happens, Copilot will show matches in the agent session logs with a link to display details of the matched code.\n- **Limited scope (CLI)**: The language model used by Copilot CLI has been trained on a large body of code but still has a limited scope and may not be able to handle certain code structures or obscure programming languages. For each language, the quality of suggestions depends on the volume and diversity of training data for that language.\n- **Inaccurate code (CLI)**: Copilot CLI may generate code that appears to be valid but may not actually be semantically or syntactically correct, or may not accurately reflect the intent of the developer. You should carefully review and test generated code, particularly when dealing with critical or sensitive applications.\n- **Security risks (CLI)**: Copilot CLI generates code and natural language based on the context of your local environment, which can potentially expose sensitive information or vulnerabilities if not used carefully. You should review all outputs generated by the agent thoroughly.\n- **Public code matches (CLI)**: Copilot CLI may generate code that is a match or near match of publicly available code, even if the \"Suggestions matching public code\" policy is set to \"Block.\"\n- **Command execution risks (CLI)**: Additional caution is required when asking or allowing Copilot CLI to execute a command, particularly regarding the potential destructiveness of some suggested commands. You may encounter commands for file deletion or hard drive formatting, which can cause problems if used incorrectly. You are ultimately responsible for the commands executed by Copilot CLI.\n- **Inherited limitations (SDK)**: Because the Copilot SDK communicates with Copilot CLI, applications built with the SDK inherit the same model limitations, including limited scope for certain programming languages and the potential for inaccurate or insecure code generation.\n- **Custom agent complexity (SDK)**: Incorrectly configured custom agents, tools, or hooks may produce unexpected behavior. Developers are responsible for testing and validating the behavior of their custom agent configurations.\n- **MCP server trust (SDK)**: MCP servers connected through the SDK can expose tools and data from external sources. Developers must ensure that connected MCP servers are trustworthy, as malicious or misconfigured servers could introduce harmful behavior or expose sensitive data.\n- **BYOK model variance (SDK)**: When using bring-your-own-key configurations with third-party model providers, behavior may differ from GitHub-hosted models. Developers are responsible for evaluating the safety and quality of responses from their chosen provider.\n\n## [8\\. Evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#8-evaluations)\n\nPerformance and safety evaluations assess whether AI applications are operating reliably and securely by examining factors like groundedness, relevance, and coherence while identifying the risks of generating harmful content. The following evaluations were conducted with safety components already in place, which are also described in [9\\. Safety components and mitigations](https://docs.github.com/en/copilot/responsible-use/agents#9-safety-components-and-mitigations).\n\n### [Performance and quality evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#performance-and-quality-evaluations)\n\nGitHub Copilot Agents are evaluated across its supported surfaces using a combination of industry-standard benchmarks (e.g., SWE-Bench) and internally developed evaluation suites. Benchmark tasks are sourced from public open-source repositories and synthetic scenarios; no real user queries or customer code are used. Each evaluation includes multiple independent runs to account for nondeterminism in model outputs. Key metrics include resolution rate (percentage of tasks successfully completed), token efficiency, latency, and tool call reliability. Models are re-evaluated when updates are made and monitored continuously in production via error rates, response latency, and aggregate usage patterns.\n\n### [Performance and quality evaluation methods](https://docs.github.com/en/copilot/responsible-use/agents\\#performance-and-quality-evaluation-methods)\n\nNew models for GitHub Copilot Agents undergo a staged evaluation process before deployment. Integrator teams run benchmark suites specific to their surface, testing the model on representative coding tasks such as bug fixes, code generation, and multi-file refactoring. Results are reviewed against established baselines and existing production models. Models must meet or exceed baseline performance across key metrics like resolution rate, token efficiency, and latency, before advancing to the next stage. A cross-functional review board makes a formal go/no-go decision before any model is approved for user-facing deployment.\n\n### [Risk and safety evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#risk-and-safety-evaluations)\n\nEvaluating potential risks associated with AI-generated content is essential for safeguarding against content risks with varying degrees of severity. This includes evaluating an AI application's predisposition towards generating harmful content or testing vulnerabilities to jailbreak attacks. For GitHub, we conduct performance evaluations, including those which are adapted for coding purposes from [Microsoft Foundry](https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/evaluation-evaluators/risk-safety-evaluators):\n\n- Hate and unfairness\n- Sexual\n- Violence\n- Self-harm\n- Protected material\n- Jailbreak\n- Code vulnerability\n\n### [Evaluation data for quality and safety](https://docs.github.com/en/copilot/responsible-use/agents\\#evaluation-data-for-quality-and-safety)\n\nOur evaluation data is custom-built to assess AI application performance across key areas of **safety** and **quality**, simulating real-world scenarios and risks. We begin by identifying relevant evaluation aspects of concern based on multi-disciplinary research and expert input. These concerns are translated into targeted evaluation objectives and guide formulation of evaluation metrics. For **safety**, we create adversarial prompts to elicit undesirable or edge-case responses, which are then scored using AI-assisted annotators trained to assess alignment with GitHub\u2019s standards. For **quality**, we craft rubric-based prompts relevant to scenarios including evaluating retrieval-augmented generation (RAG) applications and agents. Datasets are curated from diverse sources including synthetic and public datasets to simulate real-world user scenarios. Using the curated datasets, both evaluations undergo iterative refinement and human alignment to improve metric efficacy and reliability. This methodology forms the foundation of repeatable, rigorous assessments that reflect how customers use evaluations to build better AI.\n\n### [Custom evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#custom-evaluations)\n\nCopilot agentic features have been subject to RAI red teaming to identify and address potential risks. We continue to monitor the efficacy and safety of these features over time. For more information, see [Microsoft AI Red Team building future of safer AI](https://www.microsoft.com/en-us/security/blog/2023/08/07/microsoft-ai-red-team-building-future-of-safer-ai/) on the Microsoft security blog.\n\n## [9\\. Safety components and mitigations](https://docs.github.com/en/copilot/responsible-use/agents\\#9-safety-components-and-mitigations)\n\n### [Copilot cloud agent](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cloud-agent-1)\n\n- **Privilege escalation controls**: The cloud agent only responds to interactions from users with repository write access. Actions workflows triggered by pull requests raised by the agent require approval from a user with write access before they will run. The agent filters hidden characters that might allow users to hide harmful instructions in comments or issue contents.\n- **Constrained permissions**: The cloud agent only has access to the repository where it is creating a pull request and cannot access other repositories. It can only push to a single branch: the existing pull request branch when triggered via `@copilot`, or otherwise to a new `copilot/` branch. This means that Copilot cannot push directly to your default branch (for example, `main`). The agent does not have access to Actions organization or repository secrets\u2014only secrets and variables specifically added to the `copilot` environment are passed to the agent.\n- **Ensuring traceability**: The cloud agent's commits are authored by Copilot, with the human who started the task marked as the co-author. This makes it easier to identify code generated by the agent and who initiated the task. The cloud agent's commits are signed, so they appear as \"Verified\" on GitHub. This provides confidence that the commits were made by GitHub Copilot cloud agent and have not been altered. Each commit message includes a link to the agent session logs. This gives you a permanent link from any agent-authored commit to the full session logs, so you can understand why Copilot made a change during code review or trace it later for auditing purposes.\n- **Firewall for data exfiltration prevention**: By default, the cloud agent has a firewall enabled to prevent exfiltration of code or other sensitive data, either accidentally or due to malicious user input.\n- **Automated security vulnerability detection**: During code generation, the cloud agent automatically analyzes newly generated code for security vulnerabilities and attempts to resolve them. Analysis is performed using CodeQL (to identify potential vulnerabilities and errors), secret scanning (to ensure secrets aren't introduced), and dependency analysis (to check for known vulnerabilities in referenced dependencies).\n\n### [Copilot CLI](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cli-1)\n\n- **Directory-scoped access**: By default, Copilot CLI only has access to files and folders in, and below, the directory from which it was invoked. If the agent wishes to access files outside the current directory, it will ask for permission.\n- **Permission prompts for file modifications**: Copilot CLI asks for permission before modifying files. You should ensure it is modifying the correct files before granting permission.\n- **Permission prompts for command execution**: Copilot CLI asks for permission before executing commands that may be dangerous. You should review these commands carefully before giving permission to run.\n- **Configurable permissions**: You can grant Copilot CLI specific permissions, or all permissions, by using the various command line options: for example, `--allow-tool=[TOOLS...]`, `--allow-all-tools`, `--allow-all` (or its slash command equivalent `/allow-all` for use in an interactive session). For more information, see [GitHub Copilot CLI command reference](https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference#command-line-options). Typically, when you use Copilot CLI in autopilot mode, you will grant it full permissions to allow it to complete a task autonomously, without requiring you to approve activity as it works on the task. For more information, see [Allowing GitHub Copilot CLI to work autonomously](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/autopilot).\n- **Security considerations**: For more information about security practices while using Copilot CLI, see [About GitHub Copilot CLI](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-copilot-cli#security-considerations).\n\n### [Copilot SDK](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-sdk-1)\n\n- **Inherited CLI safety controls**: The Copilot SDK communicates with Copilot CLI, inheriting its permission model and safety controls. Tool executions and file modifications still require appropriate permissions.\n- **Hook-based oversight**: The SDK's lifecycle hooks (such as `onPreToolUse` and `onPostToolUse`) allow developers to implement custom safety checks, audit logging, and approval workflows before and after tool execution.\n- **MCP server isolation**: MCP servers run as separate processes. Developers can control which servers are available per session, limiting the scope of external tool access.\n- **Session scoping**: Each SDK session is isolated with its own context, tools, and permissions. Developers can control what data and capabilities are available within each session.\n- **BYOK responsibility**: When using bring-your-own-key configurations, prompts and responses are sent directly to the configured provider. Developers are responsible for reviewing the data handling policies of their chosen provider.\n\n### [GitHub Copilot app](https://docs.github.com/en/copilot/responsible-use/agents\\#github-copilot-app)\n\nThe GitHub Copilot app is a desktop application for directing agent sessions across local repositories, git worktrees, and cloud sandboxes. The app uses GitHub Copilot CLI and GitHub Copilot SDK as its foundation and adds a user interface for parallel sessions, quick chats, GitHub issue and pull request workflows, automations, and canvases.\n\n- **Workspace isolation**: App sessions can run in separate git worktrees or cloud sandboxes, helping isolate concurrent tasks from each other and from the default branch.\n- **Inherited CLI and SDK safety controls**: App sessions use GitHub Copilot CLI and GitHub Copilot SDK, so they inherit the same model limitations and safety controls for file access, file modification, command execution, session scoping, and connected tools.\n- **Human review before merge**: The app surfaces generated changes, pull request status, CI results, review activity, and session history so users can inspect work before submitting reviews, enabling agent merge, or merging pull requests.\n- **Configurable integrations and automations**: Users control which MCP servers, skills, canvas extensions, and automations are available in the app. These configurations should be reviewed before use because they can affect the tools, data, and background actions available to sessions.\n\n### [Data handling when using your own model provider (CLI)](https://docs.github.com/en/copilot/responsible-use/agents\\#data-handling-when-using-your-own-model-provider-cli)\n\nWhen you configure Copilot CLI to use your own model provider, your prompts, code context, and generated responses are sent directly to the provider you configure. They are not routed through GitHub. You are responsible for reviewing and complying with the terms of service and data handling policies of your chosen provider.\n\n#### [Telemetry](https://docs.github.com/en/copilot/responsible-use/agents\\#telemetry)\n\nWhen you use your own model provider without offline mode, Copilot CLI continues to send telemetry to GitHub as usual. This telemetry does not include your prompts or code, but it does include usage metadata.\n\nIf you enable offline mode by setting the `COPILOT_OFFLINE` environment variable to `true`, all telemetry is disabled. In offline mode, Copilot CLI only makes network requests to your configured model provider.\n\n#### [Authentication and feature availability](https://docs.github.com/en/copilot/responsible-use/agents\\#authentication-and-feature-availability)\n\nGitHub authentication is not required when using your own model provider (BYOK). Without GitHub authentication, the following features are unavailable:\n\n- `/delegate`, which hands off the session to GitHub's server-side Copilot\n- The GitHub MCP server\n- GitHub Code Search\n\nIn offline mode, web-based tools such as `web_fetch` and GitHub Code Search are also disabled.\n\n#### [No fallback to GitHub-hosted models](https://docs.github.com/en/copilot/responsible-use/agents\\#no-fallback-to-github-hosted-models)\n\nIf your model provider configuration is invalid, Copilot CLI exits with an error. It does not fall back to GitHub-hosted models. Common failures, such as connection refused, authentication errors, model not found, and timeouts, produce user-friendly messages with actionable guidance.\n\n## [10\\. Best practices for deploying and adopting GitHub Copilot agentic features](https://docs.github.com/en/copilot/responsible-use/agents\\#10-best-practices-for-deploying-and-adopting-github-copilot-agentic-features)\n\nResponsible AI is a shared commitment between GitHub and its customers. While GitHub builds AI applications with safety, fairness, and transparency at the core, customers play a critical role in deploying and using these technologies responsibly within their own contexts. To support this partnership, we offer the following best practices for deployers and end users to help customers implement responsible AI effectively.\n\n- **Exercise caution and evaluate outcomes when using Copilot agentic features for consequential decisions or in sensitive domains**:\nConsequential decisions are those that may have a legal or significant impact on a person's access to education, employment, financial platforms, government benefits, healthcare, housing, insurance, legal platforms, or that could result in physical, psychological, or financial harm. Sensitive domains\u2014such as financial platforms, healthcare, and housing\u2014require particular care due to the potential for disproportionate impact on different groups of people. When using AI for decisions in these areas, make sure that impacted stakeholders can understand how decisions are made, appeal decisions, and update any relevant input data.\n- **Evaluate legal and regulatory considerations**:\nCustomers need to evaluate potential specific legal and regulatory obligations when using any AI platforms and solutions, which may not be appropriate for use in every industry or scenario. Additionally, AI platforms or solutions are not designed for and may not be used in ways prohibited in applicable terms of service and relevant codes of conduct.\n- **Exercise human oversight when appropriate**: Human oversight is an important safeguard when interacting with AI applications. While we continuously improve our AI applications, AI might still make mistakes. The outputs generated may be inaccurate, incomplete, biased, misaligned, or irrelevant to your intended goals. This could happen due to various reasons, such as ambiguity in the inputs or limitations of the underlying models. As such, users should review the responses generated by Copilot agentic features and verify that they match their expectations and requirements.\n- **Be aware of the risk of overreliance**:\nOverreliance on AI happens when users accept incorrect or incomplete AI outputs, mainly because mistakes in AI outputs may be hard to detect. For the end-user, overreliance could result in decreased productivity, loss of trust, application abandonment, financial loss, psychological harm, physical harm, among others. (e.g. a doctor accepts an incorrect AI output).\n- **Exercise caution when designing agentic AI in sensitive domains**:\nUsers should exercise caution when designing and/or deploying agentic AI applications in sensitive domains where agent actions are irreversible or highly consequential. Additional precautions should also be taken when creating autonomous agentic AI as described further in the [GitHub Terms](https://docs.github.com/en/site-policy/github-terms).\n- **Use Copilot code review to supplement human reviews, not to replace them**: While Copilot code review can be a powerful tool for improving code quality, it is important to use it as a tool, rather than to replace human reviews. You should always review and verify the feedback generated by Copilot code review, and supplement Copilot's feedback with careful human review to ensure your code meets your requirements.\n- **Provide feedback**: If you encounter any issues or limitations with Copilot code review, we recommend that you provide feedback by using the thumbs up and thumbs down buttons on Copilot's comments. This can help GitHub improve the tool and address any concerns or limitations.\n- **Configure custom instructions**: You can configure custom instructions to help Copilot understand your coding style and best practices, improving the relevance and quality of review feedback.\n- **Ensure cloud agent tasks are well-scoped**: The more clear and well-scoped the prompt you assign to the cloud agent, the better the results. An ideal issue includes a clear description of the problem, complete acceptance criteria, and hints on what files need to be changed.\n- **Customize the cloud agent with additional context**: The cloud agent has access to semantic code search, which helps it find relevant code based on meaning rather than just exact text matches, allowing it to complete tasks faster. To further enhance performance, implement custom Copilot instructions to help the agent better understand your project and how to build, test, and validate its changes.\n- **Use the cloud agent as a tool, not a replacement**: You should always review and test the content generated by the cloud agent to ensure that it meets your requirements and is free of errors or security concerns prior to merging.\n- **Use secure coding and code review practices with the cloud agent**: Although the cloud agent can generate syntactically correct code, it may not always be secure. Follow best practices for secure coding and code review. Take the same precautions as you would with any code that uses material you did not independently originate, including rigorous testing, IP scanning, and checking for security vulnerabilities.\n- **Stay up to date**: The cloud agent is an evolving technology. Stay up to date with any new security risks or best practices that may emerge.\n- **Use Copilot CLI as a tool, not a replacement**: You should always review and verify commands and code generated by Copilot CLI to ensure they meet your requirements and are free of errors or security concerns.\n- **Review commands before execution (CLI)**: Exercise particular caution when Copilot CLI suggests executing commands, especially those that modify or delete files. You are ultimately responsible for the commands you allow the agent to run.\n- **Keep CLI tasks well-scoped**: The more clear and well-scoped the prompt you provide, the better the results. Include a clear description of the problem, acceptance criteria, and hints on what files need to be changed.\n- **Provide feedback (CLI)**: If you encounter any issues or limitations with Copilot CLI, provide feedback using the `/feedback` command.\n- **Validate custom agent behavior (SDK)**: Thoroughly test custom agents, tools, and hooks before deploying applications built with the SDK to production. Ensure that tool configurations and system prompts produce safe, expected behavior.\n- **Audit MCP server connections (SDK)**: Only connect to MCP servers that you trust. Review the tools and data that each server exposes and ensure they align with your application's security requirements.\n- **Implement safety hooks (SDK)**: Use the SDK's lifecycle hooks to implement guardrails such as content filtering, audit logging, and tool approval workflows in your applications.\n- **Scope sessions appropriately (SDK)**: Configure each SDK session with only the tools, agents, and permissions required for the task at hand. Avoid granting broad access when narrow scoping is sufficient.\n- **Review BYOK provider policies (SDK)**: If using bring-your-own-key configurations, ensure your chosen model provider's terms of service and data handling policies meet your organization's requirements.\n\n## [11\\. Learn more about GitHub Copilot agentic features](https://docs.github.com/en/copilot/responsible-use/agents\\#11-learn-more-about-github-copilot-agentic-features)\n\nFor additional guidance on the responsible use of Copilot agentic features, we recommend reviewing the following documentation:\n\n- [Using GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review)\n- [Best practices for using GitHub Copilot to work on tasks](https://docs.github.com/en/copilot/tutorials/cloud-agent/get-the-best-results)\n- [Configure the development environment](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/customize-the-agent-environment)\n- [Customizing or disabling the firewall for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-the-firewall)\n- [Build your first Copilot-powered app](https://docs.github.com/en/copilot/get-started/sdk-quickstart)\n- [Configure MCP servers for your repository](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers)\n- [About GitHub Copilot CLI](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-copilot-cli)\n- [About the GitHub Copilot app](https://docs.github.com/en/copilot/concepts/agents/github-copilot-app)\n- [GitHub Terms for Additional Products and Features](https://docs.github.com/en/site-policy/github-terms/github-terms-for-additional-products-and-features#github-copilot)\n- [Copilot Trust Center](https://copilot.github.trust.page/)\n\n### [Learn more about responsible AI](https://docs.github.com/en/copilot/responsible-use/agents\\#learn-more-about-responsible-ai)\n\n- [Microsoft AI principles](https://www.microsoft.com/en-us/ai/responsible-ai)\n- [Microsoft responsible AI resources](https://www.microsoft.com/en-us/ai/responsible-ai-resources)\n- [Microsoft Azure Learning courses on responsible AI](https://docs.microsoft.com/en-us/learn/paths/responsible-ai-business-principles/)",
      "content_chars": 48797,
      "published_date": null
    },
    {
      "rank": 6,
      "url": "https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models",
      "title": "Base and long-term support (LTS) models - GitHub Docs",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# Base and long-term support (LTS) models\n\nLearn about base models, long-term support (LTS) models, and how they affect model availability for enterprises using GitHub Copilot.\n\nCopy as Markdown\n\n## In this article\n\n## [About base models](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models\\#about-base-models)\n\nImportant\n\n- On March 18, 2026, GitHub designated GPT-5.3-Codex as the base model.\n- Base models apply only to Copilot Business and Copilot Enterprise.\n\nA base model is the default AI model that GitHub Copilot uses when no other models are enabled. The base model is automatically enabled for all Copilot Business or Copilot Enterprise accounts within 60 days after the model is designated as a base model.\n\nWhen a new model is designated a base model, the following timeline applies:\n\n| Phase | Timeline | What happens |\n| --- | --- | --- |\n| Announcement | Day 0 | GitHub announces the new base model. |\n| Upgrade window | Day 0 to Day 60 | Customers have 60 days to upgrade their IDE extensions to versions that support the new model. |\n| Enablement | Day 60 | The new model is automatically enabled on all organizations and enterprises as the base model. |\n\n## [About long-term support (LTS) models](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models\\#about-long-term-support-lts-models)\n\nImportant\n\n- On March 18, 2026, GitHub designated GPT-5.3-Codex as the LTS model.\n- LTS models apply only to Copilot Business and Copilot Enterprise.\n\nAn LTS model is an AI model that GitHub commits to supporting for an extended period of one year from its designation date. During that period, the model remains available, which allows users to build around the model without concern that it will be closing down unexpectedly.\n\n## [Further reading](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models\\#further-reading)\n\n- [Supported AI models in GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/supported-models)\n- [Models and pricing for GitHub Copilot](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing)",
      "content_chars": 2329,
      "published_date": null
    },
    {
      "rank": 7,
      "url": "https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
      "title": "Requests in GitHub Copilot (legacy)",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# Requests in GitHub Copilot (legacy)\n\nLearn about requests in Copilot, including premium requests, how they work, and how to manage your usage effectively.\n\nCopy as Markdown\n\n## In this article\n\nImportant\n\nThis article only applies to Copilot Pro and Copilot Pro+ subscribers on an existing annual plan who remained on legacy premium request-based billing after June 1, 2026.\n\n## [What is a request?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-is-a-request)\n\nA request is any interaction where you ask Copilot to do something for you\u2014whether it's generating code, answering a question, or helping you through an extension. Each time you send a prompt in a chat window or trigger a response from Copilot, you're making a request. For agentic features, only the prompts you send count as premium requests; actions Copilot takes autonomously to complete your task, such as tool calls, do not. For example, using `/plan` in Copilot CLI counts as one premium request, and any follow-up prompt you send counts as another.\n\n## [What are premium requests?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-are-premium-requests)\n\nSome Copilot features use more advanced processing power and count as premium requests. The number of premium requests a feature consumes can vary depending on the feature and the AI model used.\n\n### [Premium features](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#premium-features)\n\nThe following Copilot features can use premium requests.\n\nImportant\n\nStarting June 1, 2026, Copilot code review will have a model multiplier of 13. This means each time Copilot reviews a pull request or reviews code in your IDE, your monthly quota of Copilot premium requests will be reduced by 13.\n\n| Feature | Premium request consumption | SKU Attribution |\n| --- | --- | --- |\n| [Copilot Chat](https://docs.github.com/en/copilot/how-tos/chat-with-copilot) | Copilot Chat uses **one premium request** per user prompt, multiplied by the model's rate. This includes ask, edit, agent, and plan modes in Copilot Chat in an IDE. | Copilot premium requests |\n| [Copilot CLI](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-copilot-cli) | Each prompt to Copilot CLI uses **one premium request** with the default model. For other models, this is multiplied by the model's rate. | Copilot premium requests |\n| [Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review) | Each time Copilot reviews a pull request (when assigned as a reviewer) or reviews code in your IDE, **13 premium requests** are consumed. | Copilot premium requests |\n| [Copilot cloud agent](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-cloud-agent) | Copilot cloud agent uses **one premium request** per session, multiplied by the model's rate. A session begins when you prompt Copilot to undertake a task. In addition, each real-time steering comment made during an active session uses **one premium request** per session, multiplied by the model's rate. | Copilot cloud agent premium requests |\n| [Copilot Spaces](https://docs.github.com/en/copilot/concepts/context/spaces) | Copilot Spaces uses **one premium request** per user prompt, multiplied by the model's rate. | Copilot premium requests |\n| [OpenAI Codex Visual Studio Code integration](https://docs.github.com/en/copilot/concepts/agents/openai-codex) | While in preview, each prompt to OpenAI Codex uses **one premium request** multiplied by the model multiplier rates. | Copilot premium requests |\n| [Third-party coding agents](https://docs.github.com/en/copilot/concepts/agents/about-third-party-coding-agents) | While in preview, each prompt to a third-party coding agent uses **one premium request**, multiplied by the model's rate. | Copilot premium requests |\n\nNote\n\nPremium requests for Spark and Copilot cloud agent are tracked in dedicated SKUs from November 1, 2025. This provides better cost visibility and budget control for each AI product.\n\nTip\n\nFor instructions on viewing how many premium requests you have used and advice on how to optimize usage, see [Monitoring your GitHub Copilot usage and entitlements (legacy)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/monitor-premium-requests).\n\n## [How do request allowances work per plan?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#how-do-request-allowances-work-per-plan)\n\nNote\n\nBilling for premium requests began on June 18, 2025, for all paid Copilot plans on GitHub.com, and on August 1, 2025, on GHE.com. The request counters were only set to zero for paid plans.\n\nIf you're on a **paid plan**, you get unlimited inline suggestions. Rate limiting is in place to accommodate for high demand. See [Usage limits for GitHub Copilot](https://docs.github.com/en/copilot/concepts/usage-limits).\n\nPaid plans also receive a monthly allowance of premium requests, which can be used for advanced chat interactions, inline suggestions using premium models, and other premium features.\n\n|  | Copilot Pro | Copilot Pro+ |\n| --- | --- | --- |\n| Pricing | $10 USD per month<br>(free for some users) | $39 USD per month |\n| Premium requests | 300 per month | 1500 per month |\n| Purchase additional premium requests at $0.04/request |  |  |\n\n## [What happens to unused requests at the end of the month?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-happens-to-unused-requests-at-the-end-of-the-month)\n\nUnused requests for the previous month do not carry over to the following month. Premium request counters reset on the 1st of each month at 00:00:00 UTC. See [Monitoring your GitHub Copilot usage and entitlements (legacy)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/monitor-premium-requests).\n\n## [What if I run out of premium requests?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-if-i-run-out-of-premium-requests)\n\nNote\n\nAdditional premium requests are not available to:\n\n- Users who subscribe, or have subscribed, to Copilot Pro or Copilot Pro+ through GitHub Mobile on iOS or Android.\n\nIf you use all of your premium requests, you can still use Copilot with one of the included models for the rest of the month. This is subject to change. Response times for the included models may vary during periods of high usage. Requests to the included models may be subject to rate limiting. See [Usage limits for GitHub Copilot](https://docs.github.com/en/copilot/concepts/usage-limits).\n\nIf you need more premium requests beyond your monthly allowance set a budget for additional premium requests or upgrade to a higher plan. See [Setting up budgets to control spending on metered products](https://docs.github.com/en/billing/how-tos/set-up-budgets).\n\n## [Model multipliers](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#model-multipliers)\n\nEach model has a premium request multiplier, based on its complexity and resource usage. Your premium request allowance is deducted according to this multiplier.\n\nSee [Model multipliers for annual plans on request-based billing (legacy)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/model-multipliers-for-annual-plans).",
      "content_chars": 7875,
      "published_date": null
    },
    {
      "rank": 8,
      "url": "https://docs.github.com/copilot",
      "title": "GitHub Copilot documentation",
      "content": "[Skip to main content](https://docs.github.com/en/copilot#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# GitHub Copilot documentation\n\nYou can use GitHub Copilot to enhance your productivity and assist as you work on code.\n\n[Overview](https://docs.github.com/copilot/get-started/what-is-github-copilot) [Quickstart](https://docs.github.com/copilot/get-started/quickstart)\n\n## Articles\n\nCategory:All categories\n\n### [About agent apps](https://docs.github.com/en/copilot/concepts/agents/agent-apps)\n\nAgent apps let you use partner-built agents directly in your workflows on GitHub, powered by your Copilot subscription.\n\nLearn about CopilotAuthor and optimize with Copilot\n\n### [About agent management](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/agent-management)\n\nUse one centralized control page to jump between agent sessions, check progress, and stay in control without losing your place.\n\nLearn about Copilot\n\n### [About agent skills](https://docs.github.com/en/copilot/concepts/agents/about-agent-skills)\n\nSkills allow Copilot to perform specialized tasks.\n\nLearn about Copilot\n\n### [About cloud and local sandboxes for GitHub Copilot](https://docs.github.com/en/copilot/concepts/about-cloud-and-local-sandboxes)\n\nCloud and local sandboxes provide isolated execution environments that let Copilot safely interact with code, tools, filesystem, and network resources securely on your local machine or in fully isolated cloud environments.\n\nLearn about Copilot\n\n### [About Copilot auto model selection](https://docs.github.com/en/copilot/concepts/models/auto-model-selection)\n\nAutomatically select the best model for each task.\n\nLearn about Copilot\n\n### [About Copilot automations](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-automations)\n\nAutomations let you run Copilot cloud agent automatically, on a schedule or in response to events in a repository.\n\nLearn about CopilotIntegrate Copilot with your tools\n\n### [About Copilot integrations](https://docs.github.com/en/copilot/concepts/tools/about-copilot-integrations)\n\nIntegrate Copilot with other tools and platforms to streamline your workflow.\n\nLearn about Copilot\n\n### [About custom agents](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-custom-agents)\n\nCustom agents enhance Copilot with assistance tailored to your needs.\n\nLearn about Copilot\n\n### [About custom agents](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-custom-agents)\n\nCustom agents enhance Copilot with assistance tailored to your needs.\n\nShowing 1-9 of 438",
      "content_chars": 2624,
      "published_date": null
    },
    {
      "rank": 9,
      "url": "https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
      "title": "Quickstart for using GitHub Copilot in your IDE",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article\n\n# Quickstart for using GitHub Copilot in your IDE\n\nUse Copilot in your IDE to explain concepts, complete code, propose edits, and validate files with agent mode.\n\nCopy markdown\n\n## In this article\n\nGitHub Copilot is available in supported IDEs: Visual Studio Code, Visual Studio, JetBrains IDEs, Eclipse, and XCode. It can answer questions and explain concepts, and suggest or complete code as you work. In IDEs that support agent mode, it can also propose edits and validate files.\n\nThis quickstart shows you how to use these capabilities. It takes about ten minutes.\n\n## [Prerequisites](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#prerequisites)\n\n- **An active Copilot plan.** See [About GitHub Copilot](https://docs.github.com/en/copilot/get-started/what-is-github-copilot#get-access).\n- **The Copilot extension for your IDE.** See [Installing the GitHub Copilot extension in your environment](https://docs.github.com/en/copilot/how-tos/set-up/install-copilot-extension).\n- **Signed in to GitHub in your IDE.** If you have authentication problems, see [Troubleshooting common issues with GitHub Copilot](https://docs.github.com/en/copilot/how-tos/troubleshoot-copilot/troubleshoot-common-issues).\n\n## [Ask a question about your code](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#ask-a-question-about-your-code)\n\n1. Open an existing code file in your editor.\n\n2. Open the Copilot Chat window.\n\nThe way you do this depends on your IDE.\n   - In Visual Studio Code, press `Ctrl` + `Alt` + `I` (Windows/Linux) or `Control` + `Command` + `I` (macOS).\n   - In other IDEs, look for Copilot Chat in the menu bar or sidebar. See [Asking GitHub Copilot questions in your IDE](https://docs.github.com/en/copilot/how-tos/chat-with-copilot/chat-in-ide).\n3. Type `what does this file do`, then press `Enter`.\n\nCopilot's answer displays below your question.\n\n4. Select a line of code in the editor, then type `explain this line` in the chat window.\n\n\nCopilot can see the file you have open, so you can ask about your code instead of describing it first.\n\n## [Let Copilot work through a task](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#let-copilot-work-through-a-task)\n\nMost IDEs support **agent mode**. In agent mode, you describe what you want, and Copilot decides which files to change, makes the edits, and runs commands with your approval.\n\nTo check whether your IDE supports agent mode, see [Copilot feature matrix](https://docs.github.com/en/copilot/reference/copilot-feature-matrix?tool=ides).\n\n1. In the chat window, select **Agent**, then describe a task. For example, `Create a task manager web app with the ability to add, delete, and mark tasks as completed`.\n\n2. Press `Enter`.\n\nCopilot generates the files and code needed.\n\n3. Review the changes, then use your IDE's controls to accept them.\n\n\nWhen you're ready, save your changes, and commit when everything looks good.\n\n## [Get your first inline suggestion](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#get-your-first-inline-suggestion)\n\nGitHub Copilot can suggest code as you work and supports many programming languages. This example uses JavaScript, but you can follow along in your preferred language.\n\n1. In your IDE, create a new JavaScript ( _\\*.js_) file.\n\n2. In the JavaScript file, type the following function header.\n\n\n\n```javascript\nfunction calculateDaysBetweenDates(begin, end) {\n```\n\n\n\nCopilot will automatically suggest an entire function body in grayed text. The exact suggestion may vary.\n\n3. To accept the suggestion, press `Tab`.\n\n\nWhen you're ready, save your changes, and commit when everything looks good.\n\n## [Next steps](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#next-steps)\n\n- Customize Copilot in your IDE with custom instructions. See [Adding repository custom instructions for GitHub Copilot in your IDE](https://docs.github.com/en/copilot/how-tos/configure-custom-instructions-in-your-ide/add-repository-instructions-in-your-ide#further-reading).\n- Try the GitHub Copilot app. See [Getting started with the GitHub Copilot app](https://docs.github.com/en/copilot/get-started/quickstart-copilot-app).",
      "content_chars": 4703,
      "published_date": null
    },
    {
      "rank": 10,
      "url": "https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
      "title": "Use GitHub Copilot CLI",
      "content": "[Skip to main content](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# Use GitHub Copilot CLI\n\nUnderstand the different ways you can use GitHub Copilot in your terminal.\n\n## [Using GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/overview)\n\nLearn how to use GitHub Copilot from the command line.\n\n## [Allowing and denying tool use](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/allowing-tools)\n\nControl which tools Copilot CLI can use to avoid unintended changes.\n\n## [Use voice input with Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/voice-input)\n\nSpeak your prompts to GitHub Copilot CLI instead of typing them, using the CLI's speech-to-text feature.\n\n## [Connecting GitHub Copilot CLI to VS Code](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/connecting-vs-code)\n\nConnect Copilot CLI to VS Code to share context, trust settings, and output.\n\n## [Delegating tasks to Copilot](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/delegate-tasks-to-cca)\n\nUse autopilot mode, or the `/delegate` slash command, to get Copilot to work autonomously on your behalf.\n\n## [Browsing issues, pull requests, and gists from GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/browse-issues-prs-gists)\n\nUse the tabs in an interactive Copilot CLI session to browse issues, pull requests and gists, without leaving the terminal.\n\n## [Asking a side question in GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/ask-a-side-question)\n\nDuring a CLI session, you can ask Copilot a question without adding the prompt, or the answer, to your conversation history.\n\n## [Rolling back changes made during a GitHub Copilot CLI session](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/roll-back-changes)\n\nRewind your Copilot CLI session to a previous prompt to undo changes in conversation history, and optionally restore files.\n\n## [Working with multiple GitHub Copilot CLI sessions](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/work-with-multiple-sessions)\n\nRun several Copilot CLI sessions at once and use the sidebar to quickly move between them.\n\n## [Invoking custom agents](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/invoke-custom-agents)\n\nUse custom agents, skills, and MCP servers in Copilot CLI to extend its capabilities.\n\n## [Steering agents in GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/steer-agents)\n\nGuide Copilot during task execution to keep work on track with your intent.\n\n## [Steering a GitHub Copilot CLI session from another device](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/steer-remotely)\n\nEnable remote control for a Copilot CLI session so you can monitor progress, respond to prompts, and continue working from GitHub.com or GitHub Mobile.\n\n## [Setting an AI credit session limit in GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/set-session-limit)\n\nLimit the amount of AI credits Copilot can spend on a session to control costs and keep tasks predictable.\n\n## [Requesting a code review with GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/agentic-code-review)\n\nUse Copilot CLI to review your code changes directly from the terminal.\n\n## [Managing pull requests with the /pr command](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/manage-pull-requests)\n\nUse the `/pr` slash command to view, create, and fix pull requests directly from Copilot CLI.\n\n## [Speeding up task completion with the /fleet command](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/speed-up-task-completion)\n\nLearn how you can speed up the completion of a multi-step implementation plan by using the `/fleet` slash command.\n\n## [Using GitHub Copilot CLI session data](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/chronicle)\n\nResume previous Copilot CLI sessions, use the `/chronicle` slash command to get insights from your session history, and ask Copilot questions about your CLI usage.",
      "content_chars": 4530,
      "published_date": null
    }
  ],
  "answer_text": null,
  "citations": [],
  "raw_response": {
    "success": true,
    "data": {
      "web": [
        {
          "url": "https://docs.github.com/copilot/about-github-copilot/what-is-github-copilot",
          "title": "About Copilot - GitHub Docs",
          "description": "GitHub Copilot is an AI assistant that helps you write, understand, and ship software. Copilot is built into GitHub, it works within the development workflow ...",
          "position": 1,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/get-started/about-github-copilot#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/get-started/about-github-copilot#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this articleNext steps\n\n# About GitHub Copilot\n\nUnderstand how GitHub Copilot helps you ship software faster, how it works, and how to get access.\n\nCopy markdown\n\n## In this article\n\nGitHub Copilot is an AI assistant that helps you write, understand, and ship software. It suggests code as you type, answers questions about a codebase, reviews your changes, and works on tasks you assign it.\n\nBecause Copilot is built into GitHub, it works within the development workflow you already use:\n\n- Your code and its history.\n- The issues and pull requests you already use to track work.\n- The automations you run in your repository.\n\nA change can start as an issue, be picked up by an agent, return as a pull request, get reviewed for quality and security, and be merged.\n\nYou and Copilot work on it in the same place, with the same context, instead of moving code into separate tools and back again.\n\n## [How Copilot works](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#how-copilot-works)\n\nCopilot's capabilities and supporting features fall into four categories:\n\n**Assistive.** Copilot responds while you work, suggesting code, answering questions, or explaining code. You review and apply each suggestion.\n\n**Agentic.** You describe a goal, and Copilot can work through multiple steps. It researches a repository, proposes a plan, edits files, reviews pull requests, runs tools, and prepares tasks for your review. You remain responsible for reviewing and approving.\n\n**Customizations.** You customize Copilot's responses with instructions, reusable prompts, repository context, external tools, agent skills, and custom agents. When Copilot understands your project's structure, conventions, and tooling, it gives you better responses.\n\n**External AI agents, models, and tools.** Coding agents, models, and tools, including MCP servers from other providers, work alongside Copilot. Third-party coding agents work asynchronously on development tasks and can make changes and open or update pull requests for your review. They are subject to the same security protections, mitigations, and limitations as Copilot.\n\n## [What you can do with Copilot](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#what-you-can-do-with-copilot)\n\nCopilot helps you throughout the software development lifecycle. For example:\n\n- **Understand code** by asking questions about files, repositories, and development concepts.\n- **Plan and implement tasks** with an agent that can research a repository, make changes, and prepare a pull request for your review.\n- **Customize responses** with instructions, reusable prompts, repository context, tools, and specialized agents.\n- **Write code** with inline suggestions and natural-language prompts.\n- **Build applications** from natural-language descriptions.\n- **Change code** by asking Copilot to edit, refactor, or fix code.\n- **Review code** and pull requests, including summaries and suggestions for potential improvements.\n\nThe capabilities available to you depend on your Copilot plan, the client you use, and your organization's policies.\n\n## [Get access](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#get-access)\n\nYou can start using Copilot in several ways, depending on your role and needs.\n\n### [Individuals](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#individuals)\n\n- **Try Copilot for free.** Use [Copilot Free](https://github.com/copilot?ref_product=copilot&ref_plan=free&ref_style=text&ref_type=engagement) to explore core features with no paid plan required.\n- **Subscribe to a paid plan.** Upgrade to a paid plan for access to premium features, increased access to models, and a higher monthly allowance of AI credits. See [Getting started with a GitHub Copilot plan](https://docs.github.com/en/copilot/how-tos/manage-your-account/get-started-with-a-copilot-plan) for more information.\n- **Get free access if you're eligible.** Students, teachers, and open source maintainers may qualify for access to premium features at no cost. See [Access GitHub Copilot for free as a student](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/enable-copilot/set-up-for-students) and [Access Copilot Pro for free as a teacher or open source maintainer](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/enable-copilot/set-up-for-teachers-and-os-maintainers).\n\n### [Members of an organization](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#members-of-an-organization)\n\n- **Request access from your organization.** If your organization or enterprise has a GitHub Copilot plan, go to [https://github.com/settings/copilot](https://github.com/settings/copilot) and request access under \"Get Copilot from an organization.\"\n\n### [Organization and enterprise owners](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#organization-and-enterprise-owners)\n\nOwners of **organizations on GitHub Free or GitHub Team** can purchase Copilot Business for their organization. [Subscribe to Copilot Business](https://github.com/github-copilot/purchase?ref_product=copilot&ref_type=purchase&ref_style=button&ref_plan=business).\n\nIf your organization is owned by an enterprise that has a Copilot subscription, you can ask your enterprise owner to enable Copilot for your organization. Go to [https://github.com/settings/copilot](https://github.com/settings/copilot) and request access under \"Get Copilot from an organization.\"\n\n**Enterprise owners** can set up Copilot Business or Copilot Enterprise for their enterprise.\n\n- If you want to pay by credit card or PayPal, [subscribe here](https://github.com/github-copilot/purchase?ref_product=copilot&ref_type=purchase&ref_style=button).\n\n\n\nNote\n\n\n\nIf you are currently in a trial of GitHub Enterprise Cloud, you must convert to a paid enterprise before you can purchase GitHub Copilot.\n\n- For other payment methods, [contact GitHub's Sales team](https://github.com/enterprise/contact?ref_product=copilot&ref_type=engagement&ref_style=text).\n\n\n## [Control for organizations and enterprises](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#control-for-organizations-and-enterprises)\n\nIf you use Copilot through an organization or enterprise, administrators decide how it can be used. They control which members have access, set policies for which features are available, exclude files that Copilot shouldn't see, and review usage data and audit logs to understand how it's being used.\n\nThese settings help administrators manage Copilot across the environments and features available to their organization or enterprise. Some settings and controls depend on the Copilot plan and the client where Copilot is used. See [Administer GitHub Copilot for your team](https://docs.github.com/en/copilot/how-tos/administer-copilot).\n\n## [Next steps](https://docs.github.com/en/copilot/get-started/about-github-copilot\\#next-steps)\n\n- Start using Copilot. See [Setting up GitHub Copilot](https://docs.github.com/en/copilot/how-tos/set-up).\n- View the Copilot features available to you by navigating to your Copilot settings at [https://github.com/settings/copilot/features](https://github.com/settings/copilot/features?ref_product=copilot&ref_type=engagement&ref_style=text).",
          "metadata": {
            "og:type": "article",
            "og:site_name": "GitHub Docs",
            "og:title": "About GitHub Copilot - GitHub Docs",
            "ogUrl": "https://docs-internal.github.com/en/copilot/get-started/about-github-copilot",
            "twitter:title": "About GitHub Copilot - GitHub Docs",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-version": "free-pro-team@latest",
            "status": "200",
            "twitter:domain": "docs-internal.github.com",
            "title": "About GitHub Copilot - GitHub Docs",
            "description": "Understand how GitHub Copilot helps you ship software faster, how it works, and how to get access.",
            "language": "en",
            "viewport": "width=device-width, initial-scale=1",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "og:url": "https://docs-internal.github.com/en/copilot/get-started/about-github-copilot",
            "twitter:card": "summary",
            "twitter:url": "https://docs-internal.github.com/en/copilot/get-started/about-github-copilot",
            "twitter:description": "Understand how GitHub Copilot helps you ship software faster, how it works, and how to get access.",
            "path-product": "copilot",
            "page-document-type": "article",
            "page-content-type": "get-started",
            "ogTitle": "About GitHub Copilot - GitHub Docs",
            "ogSiteName": "GitHub Docs",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-language": "en",
            "path-article": "copilot/get-started/about-github-copilot",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-c1b89a982b79",
            "sourceURL": "https://docs.github.com/copilot/about-github-copilot/what-is-github-copilot",
            "url": "https://docs.github.com/en/copilot/get-started/about-github-copilot",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "timezone": "America/New_York",
            "proxyUsed": "basic",
            "cacheState": "miss",
            "indexId": "865c5572-b8ee-4d82-a377-57e049b9731d",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/copilot/concepts/agents/code-review",
          "title": "About GitHub Copilot code review",
          "description": "Copilot reviews your pull requests, identifies issues, and suggests fixes. This article provides an overview of Copilot code review.",
          "position": 2,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/concepts/agents/code-review#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# About GitHub Copilot code review\n\nCopilot reviews your pull requests, identifies issues, and suggests fixes you can apply in a couple of clicks.\n\n## Who can use this feature?\n\nAvailable for all paid [Copilot plans](https://github.com/features/copilot/plans?ref_product=copilot&ref_type=purchase&ref_style=text&ref_plan=cfi).\n\nCopy as Markdown\n\n## In this article\n\n## [Introduction](https://docs.github.com/en/copilot/concepts/agents/code-review\\#introduction)\n\nCopilot code review reviews code written in any language, and provides feedback. It reviews your code from multiple angles to identify issues and suggest fixes. You can apply suggested changes with a couple of clicks.\n\nThis article provides an overview of Copilot code review. To learn how to request a code review from Copilot, see [Using GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review).\n\n## [Availability](https://docs.github.com/en/copilot/concepts/agents/code-review\\#availability)\n\nCopilot code review is supported in:\n\n- GitHub.com\n- GitHub CLI\n- GitHub Mobile\n- VS Code\n- Visual Studio\n- Xcode\n- JetBrains IDEs\n- Azure DevOps (public preview)\n\nNote\n\nIf you receive Copilot from an organization, your organization must enable the **Copilot code review** option in the Copilot policy settings. This applies to reviews on GitHub.com or in GitHub Mobile. See [Managing policies and features for GitHub Copilot in your organization](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-organization/manage-policies).\n\n## [Copilot code review without a Copilot license](https://docs.github.com/en/copilot/concepts/agents/code-review\\#copilot-code-review-without-a-copilot-license)\n\nOrganization members **without a Copilot license** can use Copilot code review on GitHub.com. An enterprise administrator or organization owner must enable it. This capability is available to organizations on **Copilot Business** and **Copilot Enterprise** plans.\n\n### [Enabling code review for users without a license](https://docs.github.com/en/copilot/concepts/agents/code-review\\#enabling-code-review-for-users-without-a-license)\n\nTo allow organization members without a Copilot license to use Copilot code review, you must enable two policies:\n\n1. **AI credits paid usage**. Enable this policy first. It allows the enterprise or organization to incur charges for Copilot code review AI credits usage.\n2. **Allow members without a Copilot license to use Copilot code review in GitHub.com**. This sub-policy enables Copilot code review for users without a license.\n\nThe second policy has these characteristics:\n\n- It is disabled by default.\n- Once this policy is set at the enterprise level, it becomes **visible, but not editable** at the organization level.\n- The policy is **most restrictive**. Copilot code review is only available in repositories under an organization where you have explicitly enabled the policy.\n\n### [How it works for users without a license](https://docs.github.com/en/copilot/concepts/agents/code-review\\#how-it-works-for-users-without-a-license)\n\nWhen both policies are enabled, users without a Copilot license can request a review from Copilot code review on their pull requests in the organization's repositories.\n\nIn repositories where automatic code review is enabled, Copilot automatically reviews all pull requests. This happens regardless of whether the author has a Copilot license. For more information about how to configure automatic code review, see [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review).\n\nCopilot code review for users without a license is not available in IDEs.\n\n## [Agentic capabilities for Copilot code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#agentic-capabilities-for-copilot-code-review)\n\nCopilot code review utilizes agentic capabilities to extend its functionality.\n\n- **Full project context gathering**. This provides more specific, accurate, and contextually aware code reviews. This capability analyzes your entire repository to better understand the context of code changes.\n- **The ability to pass suggestions to Copilot cloud agent**. This automates creating a new pull request against your branch with the suggested fixes applied. Passing suggestions to Copilot cloud agent is in public preview and subject to change.\n\nThese capabilities are enabled automatically for all plans that include Copilot code review. See [Review effort level](https://docs.github.com/en/copilot/concepts/agents/code-review#review-effort-level) later in this article for information about choosing between Lite and Balanced analysis levels.\n\nIf GitHub Actions is unavailable or if Actions workflows used by Copilot code review fail, reviews will still be generated. However, they will not include the additional features provided by the agentic capabilities.\n\n### [Usage of GitHub Actions runners for agentic capabilities in code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#usage-of-github-actions-runners-for-agentic-capabilities-in-code-review)\n\nCopilot code review uses GitHub Actions to run the agentic capabilities, including full project context gathering and passing suggestions to Copilot cloud agent. By default, Copilot code review uses standard GitHub-hosted runners. You can also upgrade to larger GitHub-hosted runners for better performance, or use self-hosted runners.\n\nNote\n\nUsage of larger GitHub-hosted runners is billed at a higher per-minute rate. Self-hosted runners do not consume GitHub Actions minutes.\n\nYou do not need to have GitHub Actions enabled in your organization or enterprise to use the agentic capabilities in code review.\n\nIf your organization has disabled GitHub-hosted runners, the agentic capabilities will not be available. In this case, code reviews will fall back to a more limited review. Organizations in this situation can use self-hosted runners.\n\nFor more information on configuring runners, see [Configuring runners for GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-runners).\n\nYou can view the GitHub Actions minutes associated with Copilot code review runs. For more information, see [GitHub Actions minutes for code review](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing#pricing-and-usage-cost-considerations-for-copilot-code-review).\n\n### [Estimated consumption](https://docs.github.com/en/copilot/concepts/agents/code-review\\#estimated-consumption)\n\nA review typically consumes an estimated $0.05 USD to $1 USD worth of AI credits with \"Lite\" effort, and $0.25 USD to $5 USD worth of AI credits with \"Balanced\" effort.\n\nConsumption generally increases with pull request size and repository custom instructions, and the ranges may change as models evolve. These estimates do not include GitHub Actions minutes.\n\nRepository and organization administrators can set the default review effort level for automatic code reviews. For configuration steps, see [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review#customizing-copilot-code-review-1).\n\n## [Code review usage](https://docs.github.com/en/copilot/concepts/agents/code-review\\#code-review-usage)\n\nEach time Copilot reviews a pull request or reviews code in your IDE, the interaction consumes AI credits. The amount depends on the model used and the number of tokens processed.\n\nCode reviews have two cost components: AI credits for the model interaction (the review itself), and GitHub Actions minutes for the agentic capabilities (context gathering and tool use). For more information on GitHub Actions usage, see [Usage of GitHub Actions runners for agentic capabilities in code review](https://docs.github.com/en/copilot/concepts/agents/code-review#usage-of-github-actions-runners-for-agentic-capabilities-in-code-review).\n\nIf a repository is configured to automatically request a code review from Copilot for all new pull requests, the AI credits consumption is attributed to the pull request author. If a review is manually requested by another user, the consumption is attributed to that user instead.\n\nFor pull requests authored by Copilot cloud agent, usage is attributed first to the human co-author associated with the change. If the co-author cannot be billed, usage is billed directly to the organization.\n\nFor pull requests authored by other bots, or when a bot requests the review, usage is billed directly to the organization. These pull requests are eligible for agentic review.\n\n### [What happens when a budget is reached](https://docs.github.com/en/copilot/concepts/agents/code-review\\#what-happens-when-a-budget-is-reached)\n\nFor Copilot Business and Copilot Enterprise, code review access is governed by budget controls. If a user reaches their user-level budget, or if the enterprise or cost center spending limit is exhausted, code reviews are blocked along with other AI credits-consuming features. See [Budgets for usage-based billing](https://docs.github.com/en/copilot/concepts/billing/budgets-for-usage-based-billing#what-happens-when-a-user-is-blocked).\n\n### [Users without a Copilot license or plan that includes Copilot code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#users-without-a-copilot-license-or-plan-that-includes-copilot-code-review)\n\nUsers without access to Copilot code review do not have a monthly allowance of AI credits for it. This includes users who have no Copilot license and users on the Copilot Free plan, which does not include Copilot code review.\n\nWhen Copilot code review is enabled for these users, any AI credits they consume are billed directly to the organization or enterprise as paid additional usage. This applies to both manually requested reviews and automatic code reviews.\n\nAI credits consumed by these users are not attributed to any individual user's budget. They appear as additional usage in billing reports. Users with a Copilot license that includes code review consume AI credits from the shared pool, subject to any user-level budgets configured by their administrator.\n\n## [Model usage](https://docs.github.com/en/copilot/concepts/agents/code-review\\#model-usage)\n\nCopilot code review is a purpose-built product that uses a carefully tuned mix of models, prompts, and system behaviors to deliver consistent, high-quality feedback across a wide range of codebases. Model switching is not supported, as changing the model is likely to compromise reliability, user experience, and the quality of review comments.\n\nNote\n\nCopilot code review may use models that are not enabled on your organization's \"Models\" settings page. The \"Models\" settings page only controls Copilot Chat.\n\nSince Copilot code review is generally available, all model usage will be subject to the generally available terms. See [Managing policies and features for GitHub Copilot in your organization](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-organization/manage-policies).\n\n## [Automatic pull request reviews](https://docs.github.com/en/copilot/concepts/agents/code-review\\#automatic-pull-request-reviews)\n\nBy default, Copilot only reviews a pull request if you assign it to the pull request. However, you can configure automatic reviews.\n\n- **Individual users** on the Copilot Pro or Copilot Pro+ plan can configure Copilot to automatically review all pull requests they create.\n- **Repository owners** can configure Copilot to automatically review all pull requests in the repository that are created by people with access to Copilot.\n- **Organization owners** can configure Copilot to automatically review all pull requests in some or all of the repositories in the organization where the pull request is created by a Copilot user.\n\nIf your organization has enabled Copilot code review without a Copilot license, automatic reviews also apply to pull requests created by organization members without a license. This applies to repositories covered by a policy where automatic reviews are enabled. For more information, see [Copilot code review without a Copilot license](https://docs.github.com/en/copilot/concepts/agents/code-review#copilot-code-review-without-a-copilot-license).\n\n### [Triggering an automatic pull request review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#triggering-an-automatic-pull-request-review)\n\nThe triggers for automatic code review depend on the configuration settings.\n\n- Basic setting:\n  - When you create a pull request as an \"Open\" pull request.\n  - The first time you switch a \"Draft\" pull request to \"Open\".\n- Review new pushes:\n  - Every time you push a new commit to the pull request.\n- Review draft pull requests:\n  - Pull requests are automatically reviewed while they are still drafts, before you switch them to \"Open\".\n\nFor full instructions, see [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review).\n\nNote\n\nUnless Copilot has been configured to review each push to a pull request, it will only review a pull request once. If you make changes to the pull request after it has been automatically reviewed and you want Copilot to re-review it, you can request this manually. Click the  button next to Copilot's name in the **Reviewers** menu.\n\n## [Review effort level](https://docs.github.com/en/copilot/concepts/agents/code-review\\#review-effort-level)\n\nCopilot code review supports multiple review effort levels, so you can choose the level of thoroughness that matches the criticality of your code.\n\n- **Lite**: Standard review. Provides fast, targeted feedback on common issues such as bugs, security vulnerabilities, and style inconsistencies (default).\n- **Balanced**: Routes pull requests to a higher-reasoning model for longer analysis of complex logic, security-sensitive code, and cross-service changes. Balanced reviews use more AI credits, and may consume marginally more GitHub Actions minutes, than Lite reviews.\n\nUse Balanced for security-sensitive code, multi-service pull requests, or repositories with strict quality standards. Use Lite for routine changes where fast feedback is more important than exhaustive analysis.\n\nYou can select the review effort level when requesting a review in the pull request, under the **Reviewers** section where Copilot appears as a reviewer. Organization owners can set a default review effort level for automatic code reviews in their organization. Repository administrators can override the organization default for a specific repository.\n\nAfter Copilot code review reviews a pull request, the pull request overview comment shows the effort level used for each review run.\n\n## [Copilot approvals](https://docs.github.com/en/copilot/concepts/agents/code-review\\#copilot-approvals)\n\nNote\n\nCopilot approvals are in public preview and subject to change.\n\nEvery Copilot code review includes an approval assessment in the overview comment, indicating whether Copilot has determined the pull request ready to approve after reviewing it. By default, Copilot's reviews do not count toward required approvals for the pull request.\n\nWhen Copilot approvals are enabled in repository, organization, and enterprise settings, Copilot can submit an approving review that satisfies your repository's required-approval rule the same way a teammate's approval would. If new commits are pushed after Copilot approves, the approval is dismissed, and you can re-request a review.\n\n## [MCP servers and agent skills for code review](https://docs.github.com/en/copilot/concepts/agents/code-review\\#mcp-servers-and-agent-skills-for-code-review)\n\nCopilot code review can use repository-level agent skills and MCP servers when they are relevant to the review.\n\nCopilot code review is more likely to use skills and MCP context when your repository or pull request gives clear signals, including review-focused skill directory names, custom instructions that reference MCP context, and pull request descriptions that include identifiers referencing configured MCP servers such as issue keys or incident IDs.\n\n### [Agent skills](https://docs.github.com/en/copilot/concepts/agents/code-review\\#agent-skills)\n\nIf your repository includes agent skills, Copilot code review can automatically use relevant skills when reviewing a pull request, extending Copilot beyond its built-in analysis.\n\nWhen reviewing a pull request, Copilot reads repository custom instructions, agent instructions, and agent skills from the head branch (the branch with your changes), not the base branch. For example, when merging `my-feature-branch` into `main`, Copilot uses the instructions and skills in `my-feature-branch`, so you can test changes to them in the same pull request without merging them first.\n\nFor setup details, see [Adding agent skills for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/add-skills).\n\n### [MCP servers](https://docs.github.com/en/copilot/concepts/agents/code-review\\#mcp-servers)\n\nCopilot code review can use MCP servers to pull context directly into the review from the third-party platforms and internal systems your team uses, including issue tracking, documentation, service catalogs, and incident tooling.\n\nThe GitHub MCP server and Playwright MCP server are enabled by default.\n\nYou can configure MCP servers in your repository settings. Repository MCP configuration on GitHub applies to both Copilot cloud agent and Copilot code review. Changes you make to repository MCP settings affect both features. For setup details, see [Configure MCP servers for your repository](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers).\n\nIn repository settings, **Allow Copilot to use MCP tools when reviewing pull requests** is enabled by default. Disable this setting if you want MCP servers available only for Copilot cloud agent, and not for Copilot code review. For step-by-step instructions, see [Configure MCP servers for your repository](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers#disabling-mcp-tools-for-code-review).\n\n## [Excluded files](https://docs.github.com/en/copilot/concepts/agents/code-review\\#excluded-files)\n\nSome file types are excluded from Copilot code review:\n\n- Dependency management files, such as package.json and Gemfile.lock\n- Log files\n- SVG files\n\nIf you include these file types in a pull request, Copilot code review will not review the file.\n\nFor more information, see [Files excluded from GitHub Copilot code review](https://docs.github.com/en/copilot/reference/review-excluded-files).\n\n## [Enhancing Copilot's knowledge of a repository](https://docs.github.com/en/copilot/concepts/agents/code-review\\#enhancing-copilots-knowledge-of-a-repository)\n\nThe more Copilot knows about the code in your repository, the tools you use, and your coding standards and practices, the more accurate and useful its reviews will become. You can enhance Copilot's knowledge of your repositories in two ways.\n\n### [Custom instructions](https://docs.github.com/en/copilot/concepts/agents/code-review\\#custom-instructions)\n\nThese are short, natural-language statements that you write and store as one or more files in a repository. If you are the owner of an organization on GitHub, you can also define custom instructions in the settings for your organization. For more information, see [About customizing GitHub Copilot responses](https://docs.github.com/en/copilot/concepts/prompting/response-customization?tool=webui#about-repository-custom-instructions).\n\n### [Choosing between custom instructions, AGENTS.md, and skills](https://docs.github.com/en/copilot/concepts/agents/code-review\\#choosing-between-custom-instructions-agentsmd-and-skills)\n\nCopilot code review can draw on several sources of customization, and each serves a different purpose. Use `.github/copilot-instructions.md` for repository-wide rules specific to Copilot, use path-specific `*.instructions.md` files under `.github/instructions/` for rules that apply only to certain files or directories, use `AGENTS.md` for standing rules you want to share across AI tools and agents, and use skills for task-specific workflows that Copilot runs on demand. The following table summarizes when to use each and how to provide rules.\n\n| Use | `copilot-instructions.md` | Path-specific `*.instructions.md` | `AGENTS.md` | Skills |\n| --- | --- | --- | --- | --- |\n| Best for | Repository-wide, always-on rules for Copilot | Always-on rules for specific paths, file types, or directories | Always-on rules shared across AI agents | Task-specific review workflows |\n| Stored in | `.github/copilot-instructions.md` | `.github/instructions/**/*.instructions.md` | Repository root (`AGENTS.md`) | `.github/skills/...` |\n| Examples | Coding standards, architecture defaults, test expectations | `content/**` writing rules, `src/**` coding conventions, language-specific guidance | Shared repository conventions that should apply beyond Copilot | Reviews, releases, migrations, analysis |\n| Activation | Automatic | Automatic when the changed files match the instruction scope | Automatic (read from repository root) | Automatic when relevant (e.g. review-focused skills such as `code-review`), or on demand |\n| Scope | Repository-wide and Copilot-specific | Repository sub-paths and Copilot-specific | Cross-tool / agent-agnostic | Invoked per task |\n| Rule | \"Copilot, always know this for this repository\" | \"Copilot, always know this when working in these paths\" | \"Any agent, always know this\" | \"Do this when needed\" |\n\nFor more information, see [Adding agent skills for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/add-skills) and [Adding repository custom instructions for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/add-custom-instructions/add-repository-instructions).\n\n### [Copilot Memory (public preview)](https://docs.github.com/en/copilot/concepts/agents/code-review\\#copilot-memory-public-preview)\n\nIf you have a Copilot Pro, Copilot Pro+, or Copilot Max plan, you can enable Copilot Memory. This allows Copilot to store useful details it has learned about a repository. Copilot can then use this information when it reviews pull requests in that repository. For more information, see [About GitHub Copilot Memory](https://docs.github.com/en/copilot/concepts/agents/copilot-memory).\n\n## [Getting detailed code quality feedback across your repository](https://docs.github.com/en/copilot/concepts/agents/code-review\\#getting-detailed-code-quality-feedback-across-your-repository)\n\nGitHub Copilot code review reviews the changes in a pull request and suggests fixes. To add systematic feedback on the reliability and maintainability of your code, on pull requests and across your default branch, enable GitHub Code Quality.\n\nGitHub Code Quality complements Copilot code review by adding:\n\n- **Rules-based CodeQL-powered analysis** on pull requests and on your default branch.\n- **Test-coverage metrics** on pull requests, so you can see whether a change maintains or reduces coverage.\n- **One-click, Copilot-powered fixes**, including delegating remediation to Copilot cloud agent.\n- **Optional merge gating** with rulesets, so pull requests with unresolved rules-based findings (or that miss a coverage threshold) can be blocked from merging.\n\nFor more information, see [GitHub Code Quality](https://docs.github.com/en/code-security/concepts/code-quality/code-quality).\n\n## [Validating Copilot code reviews](https://docs.github.com/en/copilot/concepts/agents/code-review\\#validating-copilot-code-reviews)\n\nCopilot is not guaranteed to spot all problems or issues in a pull request. Sometimes it will make mistakes. Always validate Copilot's feedback carefully. Supplement Copilot's feedback with a human review.\n\nFor more information, see [Application card: GitHub Copilot Agents](https://docs.github.com/en/copilot/responsible-use/agents).\n\n## [Next steps](https://docs.github.com/en/copilot/concepts/agents/code-review\\#next-steps)\n\n- [Using GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review)\n- [Configuring code review by GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/set-up-copilot/configure-code-review)",
          "metadata": {
            "ogTitle": "About GitHub Copilot code review - GitHub Docs",
            "language": "en",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "page-content-type": "concepts",
            "og:type": "article",
            "path-language": "en",
            "path-article": "copilot/concepts/agents/code-review",
            "viewport": "width=device-width, initial-scale=1",
            "title": "About GitHub Copilot code review - GitHub Docs",
            "og:site_name": "GitHub Docs",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogUrl": "https://docs-internal.github.com/en/copilot/concepts/agents/code-review",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "twitter:card": "summary",
            "twitter:url": "https://docs-internal.github.com/en/copilot/concepts/agents/code-review",
            "page-document-type": "article",
            "ogSiteName": "GitHub Docs",
            "path-product": "copilot",
            "status": "200",
            "og:title": "About GitHub Copilot code review - GitHub Docs",
            "twitter:domain": "docs-internal.github.com",
            "og:url": "https://docs-internal.github.com/en/copilot/concepts/agents/code-review",
            "twitter:title": "About GitHub Copilot code review - GitHub Docs",
            "twitter:description": "Copilot reviews your pull requests, identifies issues, and suggests fixes you can apply in a couple of clicks.",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-version": "free-pro-team@latest",
            "description": "Copilot reviews your pull requests, identifies issues, and suggests fixes you can apply in a couple of clicks.",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-c73d5ee2727a",
            "sourceURL": "https://docs.github.com/copilot/concepts/agents/code-review",
            "url": "https://docs.github.com/en/copilot/concepts/agents/code-review",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "proxyUsed": "basic",
            "cacheState": "hit",
            "cachedAt": "2026-09-15T21:59:34.119Z",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
          "title": "About GitHub Copilot cloud agent - GitHub Enterprise Cloud Docs",
          "description": "Copilot can research a repository, create an implementation plan, and make code changes on a branch. You can review the diff, iterate, and create a pull request ...",
          "position": 3,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent#main-content)\n\n[Skip to content](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article\n\n# About GitHub Copilot cloud agent\n\nCopilot can research a repository, create an implementation plan, and make code changes on a branch. You can review the diff, iterate, and create a pull request when you're ready.\n\nCopy markdown\n\n## Who can use this feature?\n\nCopilot cloud agent is available for all paid Copilot plans.\n\nThe agent is available in all repositories stored on GitHub, except repositories owned by managed user accounts and where it has been explicitly disabled.\n\n[Sign up for Copilot](https://github.com/features/copilot/plans?ref_product=copilot&ref_type=engagement&ref_style=button)\n\n## In this article\n\n## [Overview of Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#overview-of-copilot-cloud-agent)\n\nWith Copilot cloud agent, GitHub Copilot can work independently in the background to complete tasks, just like a human developer.\n\nCopilot cloud agent can:\n\n- Research a repository\n- Create implementation plans\n- Fix bugs\n- Implement incremental new features\n- Improve test coverage\n- Update documentation\n- Address technical debt\n- Resolve merge conflicts\n\nWhen you delegate tasks to Copilot cloud agent, you can:\n\n- Use the agents panel or other agents entry points on GitHub.com to have Copilot research, plan, and make code changes on a branch, then iterate before creating a pull request. You can also specify in your prompt that you want a pull request created right away. See [Research, plan, and iterate on code changes with Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/use-copilot-agents/research-plan-iterate).\n- Ask Copilot to open a new pull request from other entry points, including GitHub Issues and Visual Studio Code. See [Starting GitHub Copilot sessions](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent/start-copilot-sessions).\n- Mention `@copilot` in a comment on an existing pull request to ask it to make changes. See [Using Copilot cloud agent on GitHub](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent/use-cloud-agent-on-github).\n- Set up an automation to run Copilot automatically, on a schedule or in response to events such as an issue being opened. See [About Copilot automations](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-automations).\n- Assign security alerts to Copilot from security campaigns. See [Fixing alerts in a security campaign](https://docs.github.com/en/enterprise-cloud@latest/code-security/how-tos/manage-security-alerts/remediate-alerts-at-scale/fixing-alerts-in-security-campaign#assigning-alerts-to-copilot-cloud-agent).\n\nCopilot cloud agent will evaluate the task it has been assigned based on the prompt you give it.\n\nWhile working on a coding task, Copilot cloud agent has access to its own ephemeral development environment, powered by GitHub Actions, where it can explore your code, make changes, execute automated tests and linters and more.\n\nNote\n\nDeep research, planning, and iterating on code changes before creating a pull request are only available with Copilot cloud agent on GitHub.com, and in public preview for the Microsoft Teams and Slack integrations. Other cloud agent integrations (such as Azure Boards, JIRA, or Linear) only support creating a pull request directly.\n\n### [Benefits over traditional AI workflows](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#benefits-over-traditional-ai-workflows)\n\nWhen used effectively, Copilot cloud agent offers productivity benefits over traditional AI assistants in IDEs:\n\n- With **AI assistants in IDEs**, coding happens **locally**. Individual developers pair in **synchronous** sessions with the AI assistant. Decisions made during the session are **untracked** and lost to time unless committed. Although the assistant helps write code, the developer still has a lot of **manual steps** to do: create the branch, write commit messages, push the changes, open the PR, write the PR description, get a review, iterate in the IDE, and repeat. These steps take time and effort that may be hard to justify for simple or routine issues.\n\n- With **Copilot cloud agent**, all coding and iterating happens **on GitHub**. You can ask Copilot to **research** a repository, **create a plan**, and **make code changes** on a branch\u2014all before opening a pull request. You can create multiple custom agents that specialize in different types of tasks. Copilot **automates** branch creation, commit message writing, and pushing. Developers let the agents **work in the background** and then chooses to **create a pull request** when ready. Working on GitHub adds **transparency**, with every step happening in a commit and being viewable in logs, and opens up **collaboration** opportunities for the entire team.\n\n\n### [Sharing context between Copilot Chat and Copilot cloud agent sessions](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#sharing-context-between-copilot-chat-and-copilot-cloud-agent-sessions)\n\nWhen you start a Copilot cloud agent session from Copilot Chat on GitHub, the session carries the context of your chat. You can continue chatting while the session runs, and ask follow-up questions about progress in the same conversation. For more information, see [Managing agent sessions](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/use-copilot-agents/manage-and-track-agents).\n\n## [Copilot cloud agent versus agent mode](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#copilot-cloud-agent-versus-agent-mode)\n\nCopilot cloud agent is distinct from the \"agent mode\" feature available in your IDE. Copilot cloud agent works autonomously in a GitHub Actions-powered environment to complete development tasks assigned through GitHub issues or GitHub Copilot Chat prompts. It can research a repository, create a plan, make code changes on a branch, and optionally open a pull request. In contrast, agent mode in your IDE makes autonomous edits directly in your local development environment. For more information about agent mode, see [Asking GitHub Copilot questions in your IDE](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/chat-with-copilot/chat-in-ide).\n\n## [Streamlining software development with Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#streamlining-software-development-with-copilot-cloud-agent)\n\nAssigning tasks to Copilot cloud agent can enhance your software development workflow.\n\nFor example, you can assign Copilot cloud agent to straightforward issues on your backlog by selecting \"Copilot\" as the assignee. This allows you to spend less time on these issues and more time on more complex or interesting work, or work that requires a high degree of creative thinking. Copilot cloud agent can work on \"nice to have\" issues that improve the quality of your codebase or product, but often remain on the backlog while you focus on more urgent work.\n\nHaving Copilot cloud agent as an additional coding resource also allows you to start tasks that you might not have otherwise started due to lack of resources. For example, you might create issues to refactor code or add more logging, and then immediately assign these to Copilot.\n\nYou can also use Copilot cloud agent to research a repository and create a plan before any code is written, helping you understand how a codebase works or agree on an approach before committing to changes. See [Research, plan, and iterate on code changes with Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/use-copilot-agents/research-plan-iterate).\n\nCopilot cloud agent can start a task, which you then pick up and continue working on yourself. By assigning the initial work to Copilot, you free up time that you would otherwise have spent doing repetitive tasks, such as setting up the scaffolding for a new project.\n\nYou can create specialized custom agents for different tasks. For example, you might create a custom agent specialized for frontend development that focuses on React components and styling, a documentation agent that excels at writing and updating technical documentation, or a testing agent that specializes in generating comprehensive unit tests. Each custom agent can be tailored with specific prompts and tools suited to its particular task.\n\n## [Measuring pull request outcomes for Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#measuring-pull-request-outcomes-for-copilot-cloud-agent)\n\nEnterprise administrators and organization owners can use Copilot usage metrics to analyze pull request outcomes for pull requests created by Copilot cloud agent.\n\nThe Copilot usage metrics APIs include pull request lifecycle metrics such as:\n\n- The total number of pull requests created and merged\n- The number of pull requests created by Copilot cloud agent that have been merged\n- Median time to merge for merged pull requests, including pull requests created by Copilot cloud agent\n\nThese metrics can help you track adoption of Copilot cloud agent and monitor changes in pull request throughput and time to merge over time. See [GitHub Copilot usage metrics](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/copilot-usage-metrics/copilot-metrics).\n\n## [Integrating Copilot cloud agent with third-party tools](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#integrating-copilot-cloud-agent-with-third-party-tools)\n\nYou can invoke Copilot cloud agent from external tools, allowing you to assign tasks to Copilot, provide context, and open pull requests without leaving your workflow.\n\nUse Copilot cloud agent in Microsoft Teams and Slack to collaborate with your team on agent-assisted work. You can @mention GitHub in channels, threads, and direct messages to work alongside teammates and Copilot on research, planning, and coding tasks. Teammates can add context, steer Copilot sessions, monitor progress, and then review the resulting artifacts.\n\nFor more information, see [About Copilot integrations](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/tools/about-copilot-integrations).\n\n## [Making Copilot cloud agent available](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#making-copilot-cloud-agent-available)\n\nBefore you can assign tasks to Copilot cloud agent, it must be enabled. Copilot cloud agent is available for all paid Copilot plans.\n\nIf you are a GitHub Copilot Business or GitHub Copilot Enterprise subscriber, an administrator must enable the relevant policy before you can use the agent.\n\nRepository owners can choose to opt out some or all repositories from Copilot cloud agent.\n\nSee [Managing access to GitHub Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/enterprise/cloud-agent-access).\n\n## [AI models for Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#ai-models-for-copilot-cloud-agent)\n\nDepending on how you start your Copilot cloud agent task, you may be able to select the model used by Copilot cloud agent. You may find that different models perform better, or provide more useful responses, depending on the type of tasks you give Copilot.\n\nSee [Changing the AI model for GitHub Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent/changing-the-ai-model).\n\n## [Enhancing Copilot cloud agent's knowledge of a repository](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#enhancing-copilot-cloud-agents-knowledge-of-a-repository)\n\nThe more Copilot cloud agent knows about the code in your repository, the tools you use, and your coding standards and practices, the more effective it will become. There are two ways you can enhance Copilot cloud agent's knowledge of a repository.\n\n- **Custom instructions**\n\nThese are short, natural\u2011language statements that you write and store as one or more files in a repository. If you are the owner of an organization on GitHub you can also define custom instructions in the settings for your organization. See [About customizing GitHub Copilot responses](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/prompting/response-customization?tool=webui#about-repository-custom-instructions).\n\n- **Copilot Memory** (public preview)\n\nIf you have a Copilot Pro, Copilot Pro+, or Copilot Max plan, you can enable Copilot Memory. This allows Copilot to store useful details it has worked out for itself about a repository. Copilot cloud agent can then use this information when it is working in that repository. See [About GitHub Copilot Memory](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/copilot-memory).\n\n\n## [Copilot cloud agent usage costs](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#copilot-cloud-agent-usage-costs)\n\nCopilot cloud agent uses GitHub Actions minutes and AI credits. The AI credits consumed depend on the model used and the number of tokens processed during the session.\n\nWithin your included GitHub Actions minutes and AI credits, you can use Copilot cloud agent without incurring additional costs. See [Usage-based billing for organizations and enterprises](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/billing/organizations-and-enterprises/usage-based-billing).\n\nCopilot code review also consumes GitHub Actions minutes on private repositories. See [GitHub Actions billing](https://docs.github.com/en/enterprise-cloud@latest/billing/concepts/product-billing/github-actions#copilot-code-review-and-github-actions-minutes).\n\n## [Customizing Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#customizing-copilot-cloud-agent)\n\nYou can customize Copilot cloud agent in a number of ways:\n\n- **Custom instructions**: Custom instructions allow you to give Copilot additional context on your project and how to build, test and validate its changes. See [Adding repository custom instructions for GitHub Copilot](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/add-custom-instructions/add-repository-instructions).\n- **Model Context Protocol (MCP) servers**: MCP servers allow you to give Copilot access to different data sources and tools. Repository MCP settings on GitHub apply to both Copilot cloud agent and Copilot code review. The GitHub MCP server and Playwright MCP server are enabled by default for both features. See [Configure MCP servers for your repository](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers).\n- **Custom agents**: Custom agents allow you to create different specialized versions of Copilot for different tasks. For example, you could customize Copilot to be an expert frontend engineer following your team's guidelines. See [About custom agents](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-custom-agents).\n- **Hooks**: Hooks allow you to execute custom shell commands at key points during agent execution, enabling you to add validation, logging, security scanning, or workflow automation. See [About hooks for GitHub Copilot](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/hooks).\n- **Skills**: Skills allow you to enhance the ability of Copilot to perform specialized tasks with instructions, scripts, and resources. See [About agent skills](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/about-agent-skills).\n\n## [Limitations of Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#limitations-of-copilot-cloud-agent)\n\nCopilot cloud agent has certain limitations in its software development workflow and compatibility with other features.\n\n### [Limitations in Copilot cloud agent's software development workflow](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#limitations-in-copilot-cloud-agents-software-development-workflow)\n\n- **Copilot can only make changes in the repository specified when you start a task**. Copilot cannot make changes across multiple repositories in one run.\n- **By default, Copilot can only access context in the repository specified when you start a task**. The Copilot MCP server is configured by default to allow Copilot to access context (for example issues and historic pull requests) in the repository where it is working. You can, however, configure broader access through repository MCP settings. See [Configure MCP servers for your repository](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers).\n- **Copilot can only work on one branch at a time** and can open exactly one pull request to address each task it is assigned.\n- **Each Copilot cloud agent session has a maximum execution time of 59 minutes**. This is a hard limit that cannot be extended or bypassed. If a task exceeds this limit, the session will time out and stop. For complex tasks that may require more time, consider breaking the work into smaller, more focused tasks. You can configure a shorter timeout using the `timeout-minutes` setting in your `copilot-setup-steps.yml` file. See [Configure the development environment](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/customize-the-agent-environment).\n\n### [Limitations in Copilot cloud agent's compatibility with other features](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#limitations-in-copilot-cloud-agents-compatibility-with-other-features)\n\n- **Copilot isn't able to comply with certain rules that may be configured for your repository**. If you have configured a ruleset or branch protection rule that isn't compatible with Copilot cloud agent, access to the agent will be blocked. For example, a rule that only allows specific commit authors can prevent Copilot cloud agent from creating or updating pull requests. If the rule is configured using rulesets, you can add Copilot as a bypass actor to enable access. See [Creating rulesets for a repository](https://docs.github.com/en/enterprise-cloud@latest/repositories/configuring-branches-and-merges-in-your-repository/managing-rulesets/creating-rulesets-for-a-repository#granting-bypass-permissions-for-your-branch-or-tag-ruleset).\n- **Copilot cloud agent only works with repositories hosted on GitHub**. If your repository is stored using a different code hosting platform, Copilot won't be able to work on it.\n\n## [Hands-on practice](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#hands-on-practice)\n\nTry the [Expand your team with Copilot cloud agent](https://github.com/skills/expand-your-team-with-copilot/?ref_product=copilot&ref_type=engagement&ref_style=text) Skills exercise for practical experience with Copilot cloud agent.\n\n## [Further reading](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent\\#further-reading)\n\n- [GitHub Copilot cloud agent](https://docs.github.com/en/enterprise-cloud@latest/copilot/how-tos/use-copilot-agents/cloud-agent) how-to articles\n- [About custom agents](https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-custom-agents)\n- [Application card: GitHub Copilot Agents](https://docs.github.com/en/enterprise-cloud@latest/copilot/responsible-use/agents)",
          "metadata": {
            "twitter:url": "https://docs-internal.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
            "language": "en",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogTitle": "About GitHub Copilot cloud agent - GitHub Enterprise Cloud Docs",
            "twitter:title": "About GitHub Copilot cloud agent - GitHub Enterprise Cloud Docs",
            "description": "Copilot can research a repository, create an implementation plan, and make code changes on a branch. You can review the diff, iterate, and create a pull request when you're ready.",
            "twitter:domain": "docs-internal.github.com",
            "twitter:description": "Copilot can research a repository, create an implementation plan, and make code changes on a branch. You can review the diff, iterate, and create a pull request when you're ready.",
            "status": "200",
            "og:site_name": "GitHub Docs",
            "path-product": "copilot",
            "page-document-type": "article",
            "title": "About GitHub Copilot cloud agent - GitHub Enterprise Cloud Docs",
            "ogUrl": "https://docs-internal.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
            "page-content-type": "concepts",
            "path-article": "copilot/concepts/agents/cloud-agent/about-cloud-agent",
            "og:type": "article",
            "viewport": "width=device-width, initial-scale=1",
            "og:url": "https://docs-internal.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
            "path-version": "enterprise-cloud@latest",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "og:title": "About GitHub Copilot cloud agent - GitHub Enterprise Cloud Docs",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogSiteName": "GitHub Docs",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "path-language": "en",
            "twitter:card": "summary",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-caf13c2d1655",
            "sourceURL": "https://docs.github.com/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
            "url": "https://docs.github.com/en/enterprise-cloud@latest/copilot/concepts/agents/cloud-agent/about-cloud-agent",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "timezone": "America/New_York",
            "proxyUsed": "basic",
            "cacheState": "miss",
            "indexId": "d396010b-356f-43d7-a798-36f140e84d6d",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/en/copilot/concepts/learning-about-new-features-and-models",
          "title": "Learning about new features and models - GitHub Docs",
          "description": "GitHub Copilot is evolving rapidly, with new features and models released regularly. Model updates are announced in the Copilot changelog.",
          "position": 4,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article\n\n# Learning about new features and models\n\nStay informed about GitHub Copilot features and models, and make confident decisions about enabling them for your enterprise.\n\nCopy markdown\n\n## In this article\n\n## Up next\n\n[Pilot a new Copilot feature or model in your enterprise](https://docs.github.com/en/copilot/tutorials/roll-out-at-scale/govern-at-scale/pilot-a-feature-or-model)\n\nGitHub Copilot is evolving rapidly, with new features and models released regularly. If you're an enterprise administrator, staying informed helps you make confident decisions about which capabilities to enable, when to adopt them, and how to manage risk across your organizations.\n\nThis article provides guidance on navigating the Copilot landscape, understanding different types of features and models, and preparing for upcoming changes.\n\n## [Keeping up to date with Copilot changes](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#keeping-up-to-date-with-copilot-changes)\n\nYou can use GitHub Docs and the GitHub Blog to keep track of new releases.\n\n### [Learning about new Copilot features](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#learning-about-new-copilot-features)\n\nTo learn about new Copilot features, we recommend monitoring these two locations:\n\n- **Copilot blog updates**: For broader news and updates about Copilot, see the [Copilot blog](https://github.blog/ai-and-ml/github-copilot/).\n- **Changelog**: For announcements about newly released and updated features, follow the [Copilot changelog](https://github.blog/changelog/?label=copilot).\n\nCopilot's capabilities and supporting features fall into four categories:\n\n| Type | Description | Examples | Considerations |\n| --- | --- | --- | --- |\n| Assistive | Respond to prompts and provide suggestions, but require human review before changes are made. | Inline suggestions, Copilot Chat, Copilot pull request summaries. | All changes require user approval. |\n| Agentic | Autonomously research, plan, and make changes on behalf of users. | Copilot cloud agent, third-party agents. | Can work autonomously, but with built-in protections. |\n| Customizations | Add context, instructions, tools, skills, and agents to tailor how Copilot responds and works. | Custom instructions, prompt files, Copilot Spaces, MCP servers, agent skills, custom agents. | Support and configuration vary by feature and surface. Organization and enterprise policies can control access to some customizations. |\n| External AI agents, models, and tools | External coding agents, models, and tools, including MCP servers from other providers, that work alongside Copilot cloud agent to complete tasks asynchronously. | Anthropic Claude, OpenAI Codex. | Same as agentic. Review provider documentation and enable via policies. See [About third-party coding agents](https://docs.github.com/en/copilot/concepts/agents/about-third-party-coding-agents). |\n\nEach feature has its own enablement requirements and policy settings. When a new feature is released:\n\n1. Review the feature documentation to understand its capabilities.\n2. Check the policy settings available at the enterprise and organization level. See [GitHub Copilot policies for enterprises and organizations](https://docs.github.com/en/copilot/concepts/enterprise/policies).\n3. Consider running a pilot with a subset of users before broader rollout.\n\n### [Learning about new Copilot models](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#learning-about-new-copilot-models)\n\nGitHub Copilot uses multiple AI models from different providers to power its features. Different models have different strengths: some prioritize speed and cost-efficiency, while others are optimized for accuracy, reasoning, or working with multimodal inputs (like images and code together).\n\nUsers will have access to different models depending on the feature they're using and your enterprise's Copilot plan and policies.\n\nYou can find information about the models available and upcoming models in the following locations:\n\n- **Supported models**: For a complete list of available models and their capabilities, see [Supported AI models in GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/supported-models).\n- **Model comparison**: To compare model capabilities side by side, see [AI model comparison](https://docs.github.com/en/copilot/reference/ai-models/model-comparison).\n- **Changelog**: Model updates are announced in the [Copilot changelog](https://github.blog/changelog/label/copilot/).\n\n#### [Default enablement](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#default-enablement)\n\nFor enterprises on a Copilot Business or Copilot Enterprise plan, a policy controls whether unconfigured generally available (GA) models are enabled or disabled by default. See [About default availability of Copilot models](https://docs.github.com/en/copilot/concepts/enterprise/default-model-availability).\n\n#### [Special categories](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#special-categories)\n\nGitHub categorizes certain types of model, allowing you to plan for model transitions and set user expectations.\n\n| Model type | Description | Why it matters |\n| --- | --- | --- |\n| Utility model | A small set of models that power background GitHub features across surfaces. | Your enterprise cannot disable these models. Monitor these to ensure they are compliant. |\n| Base model | The default model when no other models are enabled. | Automatically enabled within 60 days of designation. |\n| LTS model | A model supported for one year from designation. | Allows enterprises to build workflows around a stable model. |\n\nFor more information, see [Utility models](https://docs.github.com/en/copilot/concepts/models/utility-models) and [Base and long-term support (LTS) models](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models).\n\n## [Considering different release stages](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#considering-different-release-stages)\n\nCopilot features and models progress through different release stages. Understanding release stages helps you decide when to enable features for your organization. Each stage has different expectations for stability and support.\n\n| Stage | Availability | Stability |\n| --- | --- | --- |\n| Private preview | Limited set of customers by invitation. | May change significantly based on feedback. |\n| Public preview | All eligible customers who opt in. | Functional but may have limitations or change. |\n| Generally available (GA) | All customers. | Fully supported and recommended for production. |\n\nTip\n\nMany preview features are covered by the [GitHub Data Protection Agreement](https://gh.io/dpa), meaning you can test them without breaching compliance requirements. See [GitHub DPA-Covered Previews](https://gh.io/dpa-previews) for a list.\n\nPreview features are controlled by policy settings at the organization and enterprise level. The **Opt in to preview features** policy allows administrators to enable or disable preview features on the GitHub website specifically. See [Managing policies and features for GitHub Copilot in your enterprise](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-enterprise/manage-enterprise-policies) and [Managing policies and features for GitHub Copilot in your organization](https://docs.github.com/en/copilot/how-tos/administer-copilot/manage-for-organization/manage-policies).\n\n## [Reviewing information about Responsible AI](https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models\\#reviewing-information-about-responsible-ai)\n\nGitHub is committed to responsible AI development and provides transparency documentation for each Copilot feature.\n\nEach Copilot feature has a dedicated **responsible use** article describing its purpose, capabilities, and limitations. See [Responsible use of GitHub Copilot features](https://docs.github.com/en/copilot/responsible-use).",
          "metadata": {
            "page-document-type": "article",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "language": "en",
            "path-language": "en",
            "path-article": "copilot/concepts/enterprise/learning-about-new-features-and-models",
            "twitter:description": "Stay informed about GitHub Copilot features and models, and make confident decisions about enabling them for your enterprise.",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "og:url": "https://docs-internal.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models",
            "og:title": "Learning about new features and models - GitHub Docs",
            "og:site_name": "GitHub Docs",
            "ogSiteName": "GitHub Docs",
            "ogTitle": "Learning about new features and models - GitHub Docs",
            "path-version": "free-pro-team@latest",
            "twitter:domain": "docs-internal.github.com",
            "page-content-type": "concepts",
            "path-product": "copilot",
            "status": "200",
            "viewport": "width=device-width, initial-scale=1",
            "title": "Learning about new features and models - GitHub Docs",
            "description": "Stay informed about GitHub Copilot features and models, and make confident decisions about enabling them for your enterprise.",
            "twitter:card": "summary",
            "ogUrl": "https://docs-internal.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models",
            "og:type": "article",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "twitter:url": "https://docs-internal.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models",
            "twitter:title": "Learning about new features and models - GitHub Docs",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "docs-team-metrics": "ai-governance",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-ccead7c9de67",
            "sourceURL": "https://docs.github.com/en/copilot/concepts/learning-about-new-features-and-models",
            "url": "https://docs.github.com/en/copilot/concepts/enterprise/learning-about-new-features-and-models",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "timezone": "America/New_York",
            "proxyUsed": "basic",
            "cacheState": "miss",
            "indexId": "84dac3e2-e66b-4140-920e-813c7b3266a4",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/en/copilot/responsible-use/agents",
          "title": "Application card: GitHub Copilot Agents",
          "description": "Learn how to use GitHub Copilot agentic features responsibly by understanding their purposes, capabilities, and limitations.",
          "position": 5,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/responsible-use/agents#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/responsible-use/agents#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article8\\. Evaluations\n\n# Application card: GitHub Copilot Agents\n\nLearn how to use GitHub Copilot agentic features responsibly by understanding their purposes, capabilities, and limitations.\n\nCopy markdown\n\n## In this article\n\n## [What is an Application Card?](https://docs.github.com/en/copilot/responsible-use/agents\\#what-is-an-application-card)\n\nGitHub\u2019s application and platform cards are intended to help you understand how our AI technology works, the choices application owners can make that influence application performance and behavior, and the importance of considering the whole application, including the technology, the people, and the environment. Application cards are created for AI applications and platform cards are created for AI platform services. These resources can support the development or deployment of your own applications and can be shared with users or stakeholders impacted by them.\n\nAs part of its commitment to responsible AI, GitHub adheres to Microsoft's [six core principles](https://www.microsoft.com/en-us/ai/principles-and-approach/?msockid=3da790040c776d6f2b5485e40de56c06#ai-principles): fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability. These principles are embedded in the [Responsible AI Standard](https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/final/en-us/microsoft-brand/documents/Microsoft-Responsible-AI-Standard-General-Requirements.pdf?culture=en-us&country=us), which guides teams in designing, building, and testing AI applications. Application and Platform Cards play a key role in operationalizing these principles by offering transparency around capabilities, intended uses, and limitations. For further insight, readers are encouraged to explore Microsoft\u2019s [Responsible AI Transparency Report](https://cdn-dynmedia-1.microsoft.com/is/content/microsoftcorp/microsoft/msc/documents/presentations/CSR/Responsible-AI-Transparency-Report-2025-vertical.pdf) and [GitHub Terms](https://docs.github.com/en/site-policy/github-terms).\n\n## [1\\. Overview](https://docs.github.com/en/copilot/responsible-use/agents\\#1-overview)\n\nGitHub Copilot includes several agentic features that go beyond suggestion and conversation\u2014they can review code, take action on your behalf, and build applications. This card covers the following experiences:\n\n- **Copilot code review**: Reviews pull request diffs and metadata on GitHub.com, producing feedback comments and suggested changes.\n- **Copilot cloud agent**: An asynchronous agent on GitHub.com that can create branches, write code, and open pull requests in response to assigned issues. The cloud agent runs in an ephemeral, firewalled environment with automated security scanning.\n- **Copilot CLI**: A command-line tool that can create and modify files, execute commands, and perform multi-step tasks. All actions require explicit permission prompts and are scoped to the current directory.\n- **Copilot SDK**: A programmatic library that allows developers to build custom AI-powered applications using Copilot. The SDK communicates with Copilot CLI over JSON-RPC and supports custom agents, MCP server integrations, lifecycle hooks, and session management.\n- **GitHub Copilot app**: A desktop application for directing agent sessions across local repositories, git worktrees, and cloud sandboxes. The app uses GitHub Copilot CLI and GitHub Copilot SDK as its foundation and adds a user interface for parallel sessions, quick chats, GitHub issue and pull request workflows, automations, and canvases.\n\nThese features share common principles\u2014human oversight, review of outputs, and responsible use\u2014but differ in their execution environments, permissions, and data flows. The sections below describe each experience in context.\n\n## [2\\. Key terms](https://docs.github.com/en/copilot/responsible-use/agents\\#2-key-terms)\n\nThe following list provides a glossary of key terms related to GitHub Copilot Agents:\n\n- **Code suggestion**: A specific code change proposed by Copilot code review as part of its feedback on a pull request. Code suggestions are presented as suggested changes that can be applied with a couple of clicks.\n- **Content filtering**: A safety system that scans prompts and responses to detect and block harmful, offensive, or insecure content before it is shown to the user.\n- **Custom instructions**: Natural language descriptions of coding style and best practices that a repository maintainer can configure to guide Copilot code review's feedback. Custom instructions help Copilot understand the conventions and standards of a specific codebase.\n- **Hallucination**: A phenomenon where a language model generates output that sounds plausible but is factually incorrect, unsupported by the provided context, or entirely fabricated. In code review, hallucination can manifest as feedback that highlights problems that do not exist or are based on misunderstandings of the code.\n- **Large language model (LLM)**: A type of neural network trained on a large body of text data that can generate, analyze, and transform natural language and code. Copilot Agents use one or more LLMs to process context and produce responses.\n- **Ephemeral development environment**: A temporary, isolated compute environment created for each cloud agent session. The environment is destroyed after the session ends, ensuring that no state persists between runs.\n- **Firewall**: A network-level control enabled by default for the cloud agent that prevents outbound connections to unauthorized hosts, protecting against accidental or malicious exfiltration of code or data.\n- **AI credit**: A unit of consumption for Copilot features. Each use of Copilot code review consumes AI credits.\n- **Permission prompt**: An interactive confirmation step in Copilot CLI that asks the user to approve an action\u2014such as modifying a file, executing a command, or accessing files outside the current directory\u2014before the agent proceeds. Permission prompts are a key safety mechanism for local agentic execution.\n- **Copilot SDK**: A programmatic library (`@github/copilot-sdk`) that lets developers build custom AI-powered applications by creating sessions, sending prompts, and receiving streaming responses from Copilot. The SDK communicates with Copilot CLI over JSON-RPC.\n- **JSON-RPC**: The communication protocol used between the Copilot SDK and Copilot CLI. The SDK sends structured requests to the CLI process, which handles model interaction and tool execution.\n- **Custom agent (SDK)**: A named agent configuration within the SDK that has its own system prompt, scoped tools, and optional MCP servers. The SDK runtime can automatically delegate to sub-agents based on user intent.\n- **Hooks (SDK)**: Lifecycle callbacks in the Copilot SDK that let developers inject custom logic at specific points during a session, such as before or after tool use, on session start or end, and on error.\n\n## [3\\. Key features or capabilities](https://docs.github.com/en/copilot/responsible-use/agents\\#3-key-features-or-capabilities)\n\nThe key features and capabilities outlined here describe what GitHub Copilot Agents are designed to do and how they perform across supported tasks.\n\n- **Automated code review feedback**: When a user requests a review from Copilot, it scans the code changes plus additional relevant context and provides feedback on the code. Feedback may include natural language comments and specific code suggestions linked to particular lines and files.\n- **Customizable review guidance**: Copilot code review can be customized with custom instructions\u2014natural language descriptions of coding style and best practices\u2014so that feedback reflects a repository's conventions and standards.\n- **Autonomous pull request creation**: The Copilot cloud agent can pick up a task from an issue, a pull request comment, or Copilot Chat, create a branch, generate tailored code changes, and open a pull request. After the initial pull request is created, the agent can iterate based on your feedback and reviews.\n- **Ephemeral, firewalled execution**: While working on a task, the cloud agent has access to its own ephemeral development environment where it can make changes to code, execute automated tests, and run linters. A firewall is enabled by default to prevent data exfiltration.\n- **Automated security scanning**: During code generation, the cloud agent automatically analyzes newly generated code for security vulnerabilities using CodeQL, secret scanning, and dependency analysis, and attempts to resolve any issues before they are introduced.\n- **External integrations**: The cloud agent can receive information and context from MCP like workIQ and Microsoft 365, and external applications like Microsoft Teams, Linear, Slack, and Jira, enabling teams to assign tasks and track progress directly within their existing workflows.\n- **Local agentic execution (Copilot CLI)**: Copilot CLI provides a chat-like interface in the terminal that can autonomously create and modify files, execute commands, and perform multi-step tasks. All actions are scoped to the current directory and require explicit permission prompts before the agent modifies files or runs commands.\n\n## [4\\. Intended uses](https://docs.github.com/en/copilot/responsible-use/agents\\#4-intended-uses)\n\nGitHub Copilot Agents can be used in multiple scenarios across a variety of industries. Some examples of use cases include:\n\n- **Supplementing human code review**: Copilot code review is intended to quickly provide feedback on a developer's code, enabling developers to get code ready to merge more quickly and increasing overall code quality.\n- **Codebase maintenance**: The cloud agent can tackle security-related fixes, dependency upgrades, and targeted refactoring.\n- **Feature development**: The cloud agent can implement incremental feature requests, develop additional test suites, and create or update documentation.\n- **Prototyping new projects**: The cloud agent and Copilot CLI can greenfield new concepts, helping developers explore ideas quickly.\n- **Setting up your environment (CLI)**: Copilot CLI can run commands in your terminal to set up your local environment to work on existing projects.\n- **Finding the right command (CLI)**: Copilot CLI can suggest commands to perform tasks you're trying to complete, and explain unfamiliar commands in natural language.\n- **Building custom AI applications (SDK)**: The Copilot SDK enables developers to build applications that leverage Copilot for code generation, natural language interaction, and task automation in their own products and workflows.\n- **Multi-agent orchestration (SDK)**: Using custom agents and sub-agents, developers can build sophisticated workflows where multiple specialized agents collaborate on complex tasks, with automatic delegation based on user intent.\n- **Extending applications with external tools (SDK)**: The SDK's MCP server support allows developers to connect their applications to external data sources and services, expanding the range of tasks their agents can perform.\n\n## [5\\. Models and training data](https://docs.github.com/en/copilot/responsible-use/agents\\#5-models-and-training-data)\n\nGitHub Copilot Agents leverage a variety of AI models to power the experience that users see. For a comparison of the models available for Copilot, see [AI model comparison](https://docs.github.com/en/copilot/reference/ai-models/model-comparison). For the full list of supported models, see [Supported AI models in GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/supported-models). For information on where models are hosted, see [Hosting of models for GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/model-hosting). To learn more about the data used to train the foundation models behind GitHub Copilot Agents, refer to the linked AI model comparison above and [What data has GitHub Copilot been trained on?](https://github.com/features/copilot#faq) in the GitHub Copilot FAQ.\n\nCopilot code review is a purpose-built product that uses a carefully tuned mix of models, prompts, and system behaviors to deliver consistent, high-quality feedback across a wide range of codebases. Model switching is not supported, as changing the model is likely to compromise reliability, user experience, and the quality of review comments. Copilot code review may use models that are not enabled on your organization's \"Models\" settings page.\n\nThe Copilot cloud agent uses a large language model to reason about tasks, generate code, and leverage tools within its ephemeral development environment. The agent has been evaluated across a variety of programming languages. English is the primary supported language for prompts and responses.\n\nCopilot CLI uses a large language model to reason about tasks, generate code, modify files, and execute commands in your local terminal environment. The agent has been evaluated across a variety of programming languages. English is the primary supported language for prompts and responses.\n\nThe Copilot SDK communicates with Copilot CLI over JSON-RPC, using the same underlying models and capabilities. Applications built with the SDK use the same models available to the authenticated Copilot user or organization. Developers can also bring their own API keys (BYOK) to use custom model providers.\n\n## [6\\. Performance](https://docs.github.com/en/copilot/responsible-use/agents\\#6-performance)\n\n### [Differences by experience](https://docs.github.com/en/copilot/responsible-use/agents\\#differences-by-experience)\n\n#### [Copilot cloud agent](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cloud-agent)\n\nThe Copilot cloud agent works by using a combination of natural language processing and machine learning to understand your task and make changes in a codebase. This process can be broken down into a number of steps:\n\n1. **Prompt processing**: The task provided through an issue, pull request comment, or Copilot Chat message is combined with other relevant, contextual information to form a prompt. That prompt is sent to a large language model for processing. Inputs can take the form of plain natural language, code snippets, or images.\n2. **Language model analysis**: The prompt is passed through a large language model, which is a neural network that has been trained on a large body of data. The language model analyzes the input prompt to help the agent reason about the task and leverage necessary tools.\n3. **Response generation**: The language model generates a response based on its analysis of the prompt. This response can take the form of natural language suggestions and code suggestions.\n4. **Output formatting**: Once the agent completes its first run, it updates the pull request description with the changes it made. The agent may include supplemental information about resources it could not access and provide suggestions on steps to resolve. You may provide feedback by commenting within the pull request or explicitly mentioning the agent (`@copilot`). The agent will then resubmit that feedback for further analysis and respond with updated changes.\n\nCopilot cloud agent is intended to provide the most relevant solution for task resolution. However, it may not always provide the answer you are looking for. You are responsible for reviewing and validating responses generated by Copilot cloud agent to ensure they are accurate and appropriate.\n\n#### [Copilot code review](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-code-review)\n\nCopilot code review inspects your code and provides feedback using a combination of natural language processing and machine learning. This process can be broken down into a number of steps:\n\n1. **Input processing**: The code changes are combined with other relevant, contextual information (for example, the pull request's title and body), and any custom instructions that have been defined, to form a prompt. That prompt is sent to a large language model.\n2. **Language model analysis**: The prompt is passed through the Copilot language model, which is a neural network that has been trained on a large body of text data. The language model analyzes the input prompt.\n3. **Response generation**: The language model generates a response based on its analysis of the input prompt. This response can take the form of natural language suggestions and code suggestions.\n4. **Output formatting**: The response is presented to the user either directly in a supported editor or as a pull request review on GitHub.com, providing code feedback linked to specific lines of specific files. Where Copilot has provided a code suggestion, the suggestion is presented as a suggested change, which can be applied with a couple of clicks.\n\n#### [Copilot CLI](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cli)\n\nCopilot CLI works by using a combination of natural language processing and machine learning to understand your task and make changes in a codebase. This process can be broken down into a number of steps:\n\n1. **Input processing**: Your input is combined with relevant contextual information to form a prompt. That prompt is sent to a large language model for processing. Inputs can take the form of plain natural language, code snippets, or references to files in your terminal.\n2. **Language model analysis**: The prompt is passed through a large language model, which is a neural network that has been trained on a large body of data. The language model analyzes the input prompt to help the agent reason about the task and use the necessary tools.\n3. **Response generation**: The language model generates a response based on its analysis of the prompt. This response can take the form of natural language suggestions, code suggestions, file modifications, and command executions.\n4. **Output formatting**: The response is formatted and presented using syntax highlighting, indentation, and other formatting features. The agent may also execute commands in your local environment and create, edit, or delete files in your file system in order to complete the task. All such actions require explicit permission prompts.\n\nYou may provide feedback to the agent after it returns a response in the interactive chat window. The agent will resubmit that feedback to the language model for further analysis and return an additional response.\n\n#### [Copilot SDK](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-sdk)\n\nThe Copilot SDK provides a programmatic interface to Copilot's agentic capabilities. Applications built with the SDK follow this process:\n\n1. **Session creation**: The application creates a session with the SDK, specifying the model, system prompt, available tools, custom agents, MCP servers, and hooks. The SDK establishes a JSON-RPC connection to Copilot CLI.\n2. **Prompt submission**: User input is submitted to the session. The SDK routes the prompt (along with session context) to the CLI, which forwards it to the language model.\n3. **Agent execution**: The language model reasons about the task and may invoke tools, delegate to sub-agents, or connect to MCP servers. Lifecycle hooks fire at each stage, allowing the application to inject custom logic.\n4. **Response streaming**: Responses are streamed back to the application, which can present them in any format appropriate for its interface. The SDK provides structured events for text, tool calls, errors, and completion signals.\n\n## [7\\. Limitations](https://docs.github.com/en/copilot/responsible-use/agents\\#7-limitations)\n\nUnderstanding GitHub Copilot agentic features' limitations is crucial to determine they are used within safe and effective boundaries. While we encourage customers to leverage these features in their innovative solutions or applications, it's important to note that they were not designed for every possible scenario. We encourage users to refer to [GitHub Terms](https://docs.github.com/en/site-policy/github-terms) as well as the following considerations when choosing a use case:\n\n- **Missed code quality problems**: Copilot may not identify all of the problems that are present in code, especially where changes are large or complex. To ensure that all relevant problems are identified and corrected, Copilot code review should be supplemented with careful human code review.\n- **False positives**: Copilot code review has a risk of hallucination\u2014it may highlight problems in reviewed code that do not exist or are based on misunderstandings of the code. Comments generated by Copilot code review should be carefully reviewed and considered before taking action and making changes.\n- **Inaccurate or insecure code suggestions**: As part of its comments, Copilot code review may provide specific code suggestions. The code generated may appear to be valid but may not actually be semantically or syntactically correct, or may not correctly resolve the problem identified in the comment. In addition, code generated by Copilot may contain security vulnerabilities or other issues. You should always carefully review and test code generated by Copilot.\n- **Potential biases**: Copilot's training data is drawn from existing code repositories, which may contain biases and errors that can be perpetuated by the tool. Additionally, Copilot code review may be biased toward certain programming languages or coding styles, which can lead to suboptimal or incomplete feedback.\n- **Limited scope (cloud agent)**: The language model used by the cloud agent has been trained on a large body of code but still has a limited scope and may not be able to handle certain code structures or obscure programming languages. For each language, the quality of suggestions depends on the volume and diversity of training data for that language.\n- **Inaccurate code (cloud agent)**: The cloud agent may generate code that appears to be valid but may not actually be semantically or syntactically correct, or may not accurately reflect the intent of the developer. You should carefully review and test generated code, particularly when dealing with critical or sensitive applications.\n- **Security risks (cloud agent)**: The cloud agent generates code and natural language based on the context of an issue or comment within a repository, which can potentially expose sensitive information or vulnerabilities if not used carefully. You should review all outputs generated by the agent thoroughly prior to merging.\n- **Public code matches (cloud agent)**: The cloud agent may generate code that is a match or near match of publicly available code, even if the \"Suggestions matching public code\" policy is set to \"Block.\" If this happens, Copilot will show matches in the agent session logs with a link to display details of the matched code.\n- **Limited scope (CLI)**: The language model used by Copilot CLI has been trained on a large body of code but still has a limited scope and may not be able to handle certain code structures or obscure programming languages. For each language, the quality of suggestions depends on the volume and diversity of training data for that language.\n- **Inaccurate code (CLI)**: Copilot CLI may generate code that appears to be valid but may not actually be semantically or syntactically correct, or may not accurately reflect the intent of the developer. You should carefully review and test generated code, particularly when dealing with critical or sensitive applications.\n- **Security risks (CLI)**: Copilot CLI generates code and natural language based on the context of your local environment, which can potentially expose sensitive information or vulnerabilities if not used carefully. You should review all outputs generated by the agent thoroughly.\n- **Public code matches (CLI)**: Copilot CLI may generate code that is a match or near match of publicly available code, even if the \"Suggestions matching public code\" policy is set to \"Block.\"\n- **Command execution risks (CLI)**: Additional caution is required when asking or allowing Copilot CLI to execute a command, particularly regarding the potential destructiveness of some suggested commands. You may encounter commands for file deletion or hard drive formatting, which can cause problems if used incorrectly. You are ultimately responsible for the commands executed by Copilot CLI.\n- **Inherited limitations (SDK)**: Because the Copilot SDK communicates with Copilot CLI, applications built with the SDK inherit the same model limitations, including limited scope for certain programming languages and the potential for inaccurate or insecure code generation.\n- **Custom agent complexity (SDK)**: Incorrectly configured custom agents, tools, or hooks may produce unexpected behavior. Developers are responsible for testing and validating the behavior of their custom agent configurations.\n- **MCP server trust (SDK)**: MCP servers connected through the SDK can expose tools and data from external sources. Developers must ensure that connected MCP servers are trustworthy, as malicious or misconfigured servers could introduce harmful behavior or expose sensitive data.\n- **BYOK model variance (SDK)**: When using bring-your-own-key configurations with third-party model providers, behavior may differ from GitHub-hosted models. Developers are responsible for evaluating the safety and quality of responses from their chosen provider.\n\n## [8\\. Evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#8-evaluations)\n\nPerformance and safety evaluations assess whether AI applications are operating reliably and securely by examining factors like groundedness, relevance, and coherence while identifying the risks of generating harmful content. The following evaluations were conducted with safety components already in place, which are also described in [9\\. Safety components and mitigations](https://docs.github.com/en/copilot/responsible-use/agents#9-safety-components-and-mitigations).\n\n### [Performance and quality evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#performance-and-quality-evaluations)\n\nGitHub Copilot Agents are evaluated across its supported surfaces using a combination of industry-standard benchmarks (e.g., SWE-Bench) and internally developed evaluation suites. Benchmark tasks are sourced from public open-source repositories and synthetic scenarios; no real user queries or customer code are used. Each evaluation includes multiple independent runs to account for nondeterminism in model outputs. Key metrics include resolution rate (percentage of tasks successfully completed), token efficiency, latency, and tool call reliability. Models are re-evaluated when updates are made and monitored continuously in production via error rates, response latency, and aggregate usage patterns.\n\n### [Performance and quality evaluation methods](https://docs.github.com/en/copilot/responsible-use/agents\\#performance-and-quality-evaluation-methods)\n\nNew models for GitHub Copilot Agents undergo a staged evaluation process before deployment. Integrator teams run benchmark suites specific to their surface, testing the model on representative coding tasks such as bug fixes, code generation, and multi-file refactoring. Results are reviewed against established baselines and existing production models. Models must meet or exceed baseline performance across key metrics like resolution rate, token efficiency, and latency, before advancing to the next stage. A cross-functional review board makes a formal go/no-go decision before any model is approved for user-facing deployment.\n\n### [Risk and safety evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#risk-and-safety-evaluations)\n\nEvaluating potential risks associated with AI-generated content is essential for safeguarding against content risks with varying degrees of severity. This includes evaluating an AI application's predisposition towards generating harmful content or testing vulnerabilities to jailbreak attacks. For GitHub, we conduct performance evaluations, including those which are adapted for coding purposes from [Microsoft Foundry](https://learn.microsoft.com/en-us/azure/ai-foundry/concepts/evaluation-evaluators/risk-safety-evaluators):\n\n- Hate and unfairness\n- Sexual\n- Violence\n- Self-harm\n- Protected material\n- Jailbreak\n- Code vulnerability\n\n### [Evaluation data for quality and safety](https://docs.github.com/en/copilot/responsible-use/agents\\#evaluation-data-for-quality-and-safety)\n\nOur evaluation data is custom-built to assess AI application performance across key areas of **safety** and **quality**, simulating real-world scenarios and risks. We begin by identifying relevant evaluation aspects of concern based on multi-disciplinary research and expert input. These concerns are translated into targeted evaluation objectives and guide formulation of evaluation metrics. For **safety**, we create adversarial prompts to elicit undesirable or edge-case responses, which are then scored using AI-assisted annotators trained to assess alignment with GitHub\u2019s standards. For **quality**, we craft rubric-based prompts relevant to scenarios including evaluating retrieval-augmented generation (RAG) applications and agents. Datasets are curated from diverse sources including synthetic and public datasets to simulate real-world user scenarios. Using the curated datasets, both evaluations undergo iterative refinement and human alignment to improve metric efficacy and reliability. This methodology forms the foundation of repeatable, rigorous assessments that reflect how customers use evaluations to build better AI.\n\n### [Custom evaluations](https://docs.github.com/en/copilot/responsible-use/agents\\#custom-evaluations)\n\nCopilot agentic features have been subject to RAI red teaming to identify and address potential risks. We continue to monitor the efficacy and safety of these features over time. For more information, see [Microsoft AI Red Team building future of safer AI](https://www.microsoft.com/en-us/security/blog/2023/08/07/microsoft-ai-red-team-building-future-of-safer-ai/) on the Microsoft security blog.\n\n## [9\\. Safety components and mitigations](https://docs.github.com/en/copilot/responsible-use/agents\\#9-safety-components-and-mitigations)\n\n### [Copilot cloud agent](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cloud-agent-1)\n\n- **Privilege escalation controls**: The cloud agent only responds to interactions from users with repository write access. Actions workflows triggered by pull requests raised by the agent require approval from a user with write access before they will run. The agent filters hidden characters that might allow users to hide harmful instructions in comments or issue contents.\n- **Constrained permissions**: The cloud agent only has access to the repository where it is creating a pull request and cannot access other repositories. It can only push to a single branch: the existing pull request branch when triggered via `@copilot`, or otherwise to a new `copilot/` branch. This means that Copilot cannot push directly to your default branch (for example, `main`). The agent does not have access to Actions organization or repository secrets\u2014only secrets and variables specifically added to the `copilot` environment are passed to the agent.\n- **Ensuring traceability**: The cloud agent's commits are authored by Copilot, with the human who started the task marked as the co-author. This makes it easier to identify code generated by the agent and who initiated the task. The cloud agent's commits are signed, so they appear as \"Verified\" on GitHub. This provides confidence that the commits were made by GitHub Copilot cloud agent and have not been altered. Each commit message includes a link to the agent session logs. This gives you a permanent link from any agent-authored commit to the full session logs, so you can understand why Copilot made a change during code review or trace it later for auditing purposes.\n- **Firewall for data exfiltration prevention**: By default, the cloud agent has a firewall enabled to prevent exfiltration of code or other sensitive data, either accidentally or due to malicious user input.\n- **Automated security vulnerability detection**: During code generation, the cloud agent automatically analyzes newly generated code for security vulnerabilities and attempts to resolve them. Analysis is performed using CodeQL (to identify potential vulnerabilities and errors), secret scanning (to ensure secrets aren't introduced), and dependency analysis (to check for known vulnerabilities in referenced dependencies).\n\n### [Copilot CLI](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-cli-1)\n\n- **Directory-scoped access**: By default, Copilot CLI only has access to files and folders in, and below, the directory from which it was invoked. If the agent wishes to access files outside the current directory, it will ask for permission.\n- **Permission prompts for file modifications**: Copilot CLI asks for permission before modifying files. You should ensure it is modifying the correct files before granting permission.\n- **Permission prompts for command execution**: Copilot CLI asks for permission before executing commands that may be dangerous. You should review these commands carefully before giving permission to run.\n- **Configurable permissions**: You can grant Copilot CLI specific permissions, or all permissions, by using the various command line options: for example, `--allow-tool=[TOOLS...]`, `--allow-all-tools`, `--allow-all` (or its slash command equivalent `/allow-all` for use in an interactive session). For more information, see [GitHub Copilot CLI command reference](https://docs.github.com/en/copilot/reference/copilot-cli-reference/cli-command-reference#command-line-options). Typically, when you use Copilot CLI in autopilot mode, you will grant it full permissions to allow it to complete a task autonomously, without requiring you to approve activity as it works on the task. For more information, see [Allowing GitHub Copilot CLI to work autonomously](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/autopilot).\n- **Security considerations**: For more information about security practices while using Copilot CLI, see [About GitHub Copilot CLI](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-copilot-cli#security-considerations).\n\n### [Copilot SDK](https://docs.github.com/en/copilot/responsible-use/agents\\#copilot-sdk-1)\n\n- **Inherited CLI safety controls**: The Copilot SDK communicates with Copilot CLI, inheriting its permission model and safety controls. Tool executions and file modifications still require appropriate permissions.\n- **Hook-based oversight**: The SDK's lifecycle hooks (such as `onPreToolUse` and `onPostToolUse`) allow developers to implement custom safety checks, audit logging, and approval workflows before and after tool execution.\n- **MCP server isolation**: MCP servers run as separate processes. Developers can control which servers are available per session, limiting the scope of external tool access.\n- **Session scoping**: Each SDK session is isolated with its own context, tools, and permissions. Developers can control what data and capabilities are available within each session.\n- **BYOK responsibility**: When using bring-your-own-key configurations, prompts and responses are sent directly to the configured provider. Developers are responsible for reviewing the data handling policies of their chosen provider.\n\n### [GitHub Copilot app](https://docs.github.com/en/copilot/responsible-use/agents\\#github-copilot-app)\n\nThe GitHub Copilot app is a desktop application for directing agent sessions across local repositories, git worktrees, and cloud sandboxes. The app uses GitHub Copilot CLI and GitHub Copilot SDK as its foundation and adds a user interface for parallel sessions, quick chats, GitHub issue and pull request workflows, automations, and canvases.\n\n- **Workspace isolation**: App sessions can run in separate git worktrees or cloud sandboxes, helping isolate concurrent tasks from each other and from the default branch.\n- **Inherited CLI and SDK safety controls**: App sessions use GitHub Copilot CLI and GitHub Copilot SDK, so they inherit the same model limitations and safety controls for file access, file modification, command execution, session scoping, and connected tools.\n- **Human review before merge**: The app surfaces generated changes, pull request status, CI results, review activity, and session history so users can inspect work before submitting reviews, enabling agent merge, or merging pull requests.\n- **Configurable integrations and automations**: Users control which MCP servers, skills, canvas extensions, and automations are available in the app. These configurations should be reviewed before use because they can affect the tools, data, and background actions available to sessions.\n\n### [Data handling when using your own model provider (CLI)](https://docs.github.com/en/copilot/responsible-use/agents\\#data-handling-when-using-your-own-model-provider-cli)\n\nWhen you configure Copilot CLI to use your own model provider, your prompts, code context, and generated responses are sent directly to the provider you configure. They are not routed through GitHub. You are responsible for reviewing and complying with the terms of service and data handling policies of your chosen provider.\n\n#### [Telemetry](https://docs.github.com/en/copilot/responsible-use/agents\\#telemetry)\n\nWhen you use your own model provider without offline mode, Copilot CLI continues to send telemetry to GitHub as usual. This telemetry does not include your prompts or code, but it does include usage metadata.\n\nIf you enable offline mode by setting the `COPILOT_OFFLINE` environment variable to `true`, all telemetry is disabled. In offline mode, Copilot CLI only makes network requests to your configured model provider.\n\n#### [Authentication and feature availability](https://docs.github.com/en/copilot/responsible-use/agents\\#authentication-and-feature-availability)\n\nGitHub authentication is not required when using your own model provider (BYOK). Without GitHub authentication, the following features are unavailable:\n\n- `/delegate`, which hands off the session to GitHub's server-side Copilot\n- The GitHub MCP server\n- GitHub Code Search\n\nIn offline mode, web-based tools such as `web_fetch` and GitHub Code Search are also disabled.\n\n#### [No fallback to GitHub-hosted models](https://docs.github.com/en/copilot/responsible-use/agents\\#no-fallback-to-github-hosted-models)\n\nIf your model provider configuration is invalid, Copilot CLI exits with an error. It does not fall back to GitHub-hosted models. Common failures, such as connection refused, authentication errors, model not found, and timeouts, produce user-friendly messages with actionable guidance.\n\n## [10\\. Best practices for deploying and adopting GitHub Copilot agentic features](https://docs.github.com/en/copilot/responsible-use/agents\\#10-best-practices-for-deploying-and-adopting-github-copilot-agentic-features)\n\nResponsible AI is a shared commitment between GitHub and its customers. While GitHub builds AI applications with safety, fairness, and transparency at the core, customers play a critical role in deploying and using these technologies responsibly within their own contexts. To support this partnership, we offer the following best practices for deployers and end users to help customers implement responsible AI effectively.\n\n- **Exercise caution and evaluate outcomes when using Copilot agentic features for consequential decisions or in sensitive domains**:\nConsequential decisions are those that may have a legal or significant impact on a person's access to education, employment, financial platforms, government benefits, healthcare, housing, insurance, legal platforms, or that could result in physical, psychological, or financial harm. Sensitive domains\u2014such as financial platforms, healthcare, and housing\u2014require particular care due to the potential for disproportionate impact on different groups of people. When using AI for decisions in these areas, make sure that impacted stakeholders can understand how decisions are made, appeal decisions, and update any relevant input data.\n- **Evaluate legal and regulatory considerations**:\nCustomers need to evaluate potential specific legal and regulatory obligations when using any AI platforms and solutions, which may not be appropriate for use in every industry or scenario. Additionally, AI platforms or solutions are not designed for and may not be used in ways prohibited in applicable terms of service and relevant codes of conduct.\n- **Exercise human oversight when appropriate**: Human oversight is an important safeguard when interacting with AI applications. While we continuously improve our AI applications, AI might still make mistakes. The outputs generated may be inaccurate, incomplete, biased, misaligned, or irrelevant to your intended goals. This could happen due to various reasons, such as ambiguity in the inputs or limitations of the underlying models. As such, users should review the responses generated by Copilot agentic features and verify that they match their expectations and requirements.\n- **Be aware of the risk of overreliance**:\nOverreliance on AI happens when users accept incorrect or incomplete AI outputs, mainly because mistakes in AI outputs may be hard to detect. For the end-user, overreliance could result in decreased productivity, loss of trust, application abandonment, financial loss, psychological harm, physical harm, among others. (e.g. a doctor accepts an incorrect AI output).\n- **Exercise caution when designing agentic AI in sensitive domains**:\nUsers should exercise caution when designing and/or deploying agentic AI applications in sensitive domains where agent actions are irreversible or highly consequential. Additional precautions should also be taken when creating autonomous agentic AI as described further in the [GitHub Terms](https://docs.github.com/en/site-policy/github-terms).\n- **Use Copilot code review to supplement human reviews, not to replace them**: While Copilot code review can be a powerful tool for improving code quality, it is important to use it as a tool, rather than to replace human reviews. You should always review and verify the feedback generated by Copilot code review, and supplement Copilot's feedback with careful human review to ensure your code meets your requirements.\n- **Provide feedback**: If you encounter any issues or limitations with Copilot code review, we recommend that you provide feedback by using the thumbs up and thumbs down buttons on Copilot's comments. This can help GitHub improve the tool and address any concerns or limitations.\n- **Configure custom instructions**: You can configure custom instructions to help Copilot understand your coding style and best practices, improving the relevance and quality of review feedback.\n- **Ensure cloud agent tasks are well-scoped**: The more clear and well-scoped the prompt you assign to the cloud agent, the better the results. An ideal issue includes a clear description of the problem, complete acceptance criteria, and hints on what files need to be changed.\n- **Customize the cloud agent with additional context**: The cloud agent has access to semantic code search, which helps it find relevant code based on meaning rather than just exact text matches, allowing it to complete tasks faster. To further enhance performance, implement custom Copilot instructions to help the agent better understand your project and how to build, test, and validate its changes.\n- **Use the cloud agent as a tool, not a replacement**: You should always review and test the content generated by the cloud agent to ensure that it meets your requirements and is free of errors or security concerns prior to merging.\n- **Use secure coding and code review practices with the cloud agent**: Although the cloud agent can generate syntactically correct code, it may not always be secure. Follow best practices for secure coding and code review. Take the same precautions as you would with any code that uses material you did not independently originate, including rigorous testing, IP scanning, and checking for security vulnerabilities.\n- **Stay up to date**: The cloud agent is an evolving technology. Stay up to date with any new security risks or best practices that may emerge.\n- **Use Copilot CLI as a tool, not a replacement**: You should always review and verify commands and code generated by Copilot CLI to ensure they meet your requirements and are free of errors or security concerns.\n- **Review commands before execution (CLI)**: Exercise particular caution when Copilot CLI suggests executing commands, especially those that modify or delete files. You are ultimately responsible for the commands you allow the agent to run.\n- **Keep CLI tasks well-scoped**: The more clear and well-scoped the prompt you provide, the better the results. Include a clear description of the problem, acceptance criteria, and hints on what files need to be changed.\n- **Provide feedback (CLI)**: If you encounter any issues or limitations with Copilot CLI, provide feedback using the `/feedback` command.\n- **Validate custom agent behavior (SDK)**: Thoroughly test custom agents, tools, and hooks before deploying applications built with the SDK to production. Ensure that tool configurations and system prompts produce safe, expected behavior.\n- **Audit MCP server connections (SDK)**: Only connect to MCP servers that you trust. Review the tools and data that each server exposes and ensure they align with your application's security requirements.\n- **Implement safety hooks (SDK)**: Use the SDK's lifecycle hooks to implement guardrails such as content filtering, audit logging, and tool approval workflows in your applications.\n- **Scope sessions appropriately (SDK)**: Configure each SDK session with only the tools, agents, and permissions required for the task at hand. Avoid granting broad access when narrow scoping is sufficient.\n- **Review BYOK provider policies (SDK)**: If using bring-your-own-key configurations, ensure your chosen model provider's terms of service and data handling policies meet your organization's requirements.\n\n## [11\\. Learn more about GitHub Copilot agentic features](https://docs.github.com/en/copilot/responsible-use/agents\\#11-learn-more-about-github-copilot-agentic-features)\n\nFor additional guidance on the responsible use of Copilot agentic features, we recommend reviewing the following documentation:\n\n- [Using GitHub Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review)\n- [Best practices for using GitHub Copilot to work on tasks](https://docs.github.com/en/copilot/tutorials/cloud-agent/get-the-best-results)\n- [Configure the development environment](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-cloud-agent/customize-the-agent-environment)\n- [Customizing or disabling the firewall for GitHub Copilot](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/customize-the-firewall)\n- [Build your first Copilot-powered app](https://docs.github.com/en/copilot/get-started/sdk-quickstart)\n- [Configure MCP servers for your repository](https://docs.github.com/en/copilot/how-tos/copilot-on-github/customize-copilot/configure-mcp-servers)\n- [About GitHub Copilot CLI](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-copilot-cli)\n- [About the GitHub Copilot app](https://docs.github.com/en/copilot/concepts/agents/github-copilot-app)\n- [GitHub Terms for Additional Products and Features](https://docs.github.com/en/site-policy/github-terms/github-terms-for-additional-products-and-features#github-copilot)\n- [Copilot Trust Center](https://copilot.github.trust.page/)\n\n### [Learn more about responsible AI](https://docs.github.com/en/copilot/responsible-use/agents\\#learn-more-about-responsible-ai)\n\n- [Microsoft AI principles](https://www.microsoft.com/en-us/ai/responsible-ai)\n- [Microsoft responsible AI resources](https://www.microsoft.com/en-us/ai/responsible-ai-resources)\n- [Microsoft Azure Learning courses on responsible AI](https://docs.microsoft.com/en-us/learn/paths/responsible-ai-business-principles/)",
          "metadata": {
            "og:url": "https://docs-internal.github.com/en/copilot/responsible-use/agents",
            "twitter:title": "Application card: GitHub Copilot Agents - GitHub Docs",
            "path-article": "copilot/responsible-use/agents",
            "og:site_name": "GitHub Docs",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-language": "en",
            "ogTitle": "Application card: GitHub Copilot Agents - GitHub Docs",
            "page-content-type": "rai",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "path-version": "free-pro-team@latest",
            "ogSiteName": "GitHub Docs",
            "ogUrl": "https://docs-internal.github.com/en/copilot/responsible-use/agents",
            "viewport": "width=device-width, initial-scale=1",
            "page-document-type": "article",
            "status": "200",
            "og:type": "article",
            "language": "en",
            "og:title": "Application card: GitHub Copilot Agents - GitHub Docs",
            "title": "Application card: GitHub Copilot Agents - GitHub Docs",
            "description": "Learn how to use GitHub Copilot agentic features responsibly by understanding their purposes, capabilities, and limitations.",
            "path-product": "copilot",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "twitter:card": "summary",
            "twitter:domain": "docs-internal.github.com",
            "twitter:url": "https://docs-internal.github.com/en/copilot/responsible-use/agents",
            "twitter:description": "Learn how to use GitHub Copilot agentic features responsibly by understanding their purposes, capabilities, and limitations.",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-d17a2bf05b76",
            "sourceURL": "https://docs.github.com/en/copilot/responsible-use/agents",
            "url": "https://docs.github.com/en/copilot/responsible-use/agents",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "proxyUsed": "basic",
            "cacheState": "hit",
            "cachedAt": "2026-09-16T05:11:13.248Z",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models",
          "title": "Base and long-term support (LTS) models - GitHub Docs",
          "description": "On March 18, 2026, GitHub designated GPT-5.3-Codex as the LTS model. LTS models apply only to Copilot Business and Copilot Enterprise. An LTS model is an AI ...",
          "position": 6,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# Base and long-term support (LTS) models\n\nLearn about base models, long-term support (LTS) models, and how they affect model availability for enterprises using GitHub Copilot.\n\nCopy as Markdown\n\n## In this article\n\n## [About base models](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models\\#about-base-models)\n\nImportant\n\n- On March 18, 2026, GitHub designated GPT-5.3-Codex as the base model.\n- Base models apply only to Copilot Business and Copilot Enterprise.\n\nA base model is the default AI model that GitHub Copilot uses when no other models are enabled. The base model is automatically enabled for all Copilot Business or Copilot Enterprise accounts within 60 days after the model is designated as a base model.\n\nWhen a new model is designated a base model, the following timeline applies:\n\n| Phase | Timeline | What happens |\n| --- | --- | --- |\n| Announcement | Day 0 | GitHub announces the new base model. |\n| Upgrade window | Day 0 to Day 60 | Customers have 60 days to upgrade their IDE extensions to versions that support the new model. |\n| Enablement | Day 60 | The new model is automatically enabled on all organizations and enterprises as the base model. |\n\n## [About long-term support (LTS) models](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models\\#about-long-term-support-lts-models)\n\nImportant\n\n- On March 18, 2026, GitHub designated GPT-5.3-Codex as the LTS model.\n- LTS models apply only to Copilot Business and Copilot Enterprise.\n\nAn LTS model is an AI model that GitHub commits to supporting for an extended period of one year from its designation date. During that period, the model remains available, which allows users to build around the model without concern that it will be closing down unexpectedly.\n\n## [Further reading](https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models\\#further-reading)\n\n- [Supported AI models in GitHub Copilot](https://docs.github.com/en/copilot/reference/ai-models/supported-models)\n- [Models and pricing for GitHub Copilot](https://docs.github.com/en/copilot/reference/copilot-billing/models-and-pricing)",
          "metadata": {
            "title": "Base and long-term support (LTS) models - GitHub Docs",
            "ogTitle": "Base and long-term support (LTS) models - GitHub Docs",
            "page-document-type": "article",
            "og:title": "Base and long-term support (LTS) models - GitHub Docs",
            "og:type": "article",
            "status": "200",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "og:url": "https://docs-internal.github.com/en/copilot/concepts/models/fallback-and-lts-models",
            "page-content-type": "concepts",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "twitter:domain": "docs-internal.github.com",
            "twitter:url": "https://docs-internal.github.com/en/copilot/concepts/models/fallback-and-lts-models",
            "description": "Learn about base models, long-term support (LTS) models, and how they affect model availability for enterprises using GitHub Copilot.",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "twitter:title": "Base and long-term support (LTS) models - GitHub Docs",
            "language": "en",
            "ogUrl": "https://docs-internal.github.com/en/copilot/concepts/models/fallback-and-lts-models",
            "path-version": "free-pro-team@latest",
            "twitter:card": "summary",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-article": "copilot/concepts/models/fallback-and-lts-models",
            "ogSiteName": "GitHub Docs",
            "viewport": "width=device-width, initial-scale=1",
            "path-language": "en",
            "path-product": "copilot",
            "og:site_name": "GitHub Docs",
            "twitter:description": "Learn about base models, long-term support (LTS) models, and how they affect model availability for enterprises using GitHub Copilot.",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-d41ba19acb15",
            "sourceURL": "https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models",
            "url": "https://docs.github.com/en/copilot/concepts/models/fallback-and-lts-models",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "proxyUsed": "basic",
            "cacheState": "hit",
            "cachedAt": "2026-09-13T13:23:15.496Z",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
          "title": "Requests in GitHub Copilot (legacy)",
          "description": "Starting June 1, 2026, Premium requests for Spark and Copilot cloud agent are tracked in dedicated SKUs from November 1, 2025.",
          "position": 7,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# Requests in GitHub Copilot (legacy)\n\nLearn about requests in Copilot, including premium requests, how they work, and how to manage your usage effectively.\n\nCopy as Markdown\n\n## In this article\n\nImportant\n\nThis article only applies to Copilot Pro and Copilot Pro+ subscribers on an existing annual plan who remained on legacy premium request-based billing after June 1, 2026.\n\n## [What is a request?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-is-a-request)\n\nA request is any interaction where you ask Copilot to do something for you\u2014whether it's generating code, answering a question, or helping you through an extension. Each time you send a prompt in a chat window or trigger a response from Copilot, you're making a request. For agentic features, only the prompts you send count as premium requests; actions Copilot takes autonomously to complete your task, such as tool calls, do not. For example, using `/plan` in Copilot CLI counts as one premium request, and any follow-up prompt you send counts as another.\n\n## [What are premium requests?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-are-premium-requests)\n\nSome Copilot features use more advanced processing power and count as premium requests. The number of premium requests a feature consumes can vary depending on the feature and the AI model used.\n\n### [Premium features](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#premium-features)\n\nThe following Copilot features can use premium requests.\n\nImportant\n\nStarting June 1, 2026, Copilot code review will have a model multiplier of 13. This means each time Copilot reviews a pull request or reviews code in your IDE, your monthly quota of Copilot premium requests will be reduced by 13.\n\n| Feature | Premium request consumption | SKU Attribution |\n| --- | --- | --- |\n| [Copilot Chat](https://docs.github.com/en/copilot/how-tos/chat-with-copilot) | Copilot Chat uses **one premium request** per user prompt, multiplied by the model's rate. This includes ask, edit, agent, and plan modes in Copilot Chat in an IDE. | Copilot premium requests |\n| [Copilot CLI](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-copilot-cli) | Each prompt to Copilot CLI uses **one premium request** with the default model. For other models, this is multiplied by the model's rate. | Copilot premium requests |\n| [Copilot code review](https://docs.github.com/en/copilot/how-tos/use-copilot-agents/request-a-code-review/use-code-review) | Each time Copilot reviews a pull request (when assigned as a reviewer) or reviews code in your IDE, **13 premium requests** are consumed. | Copilot premium requests |\n| [Copilot cloud agent](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-cloud-agent) | Copilot cloud agent uses **one premium request** per session, multiplied by the model's rate. A session begins when you prompt Copilot to undertake a task. In addition, each real-time steering comment made during an active session uses **one premium request** per session, multiplied by the model's rate. | Copilot cloud agent premium requests |\n| [Copilot Spaces](https://docs.github.com/en/copilot/concepts/context/spaces) | Copilot Spaces uses **one premium request** per user prompt, multiplied by the model's rate. | Copilot premium requests |\n| [OpenAI Codex Visual Studio Code integration](https://docs.github.com/en/copilot/concepts/agents/openai-codex) | While in preview, each prompt to OpenAI Codex uses **one premium request** multiplied by the model multiplier rates. | Copilot premium requests |\n| [Third-party coding agents](https://docs.github.com/en/copilot/concepts/agents/about-third-party-coding-agents) | While in preview, each prompt to a third-party coding agent uses **one premium request**, multiplied by the model's rate. | Copilot premium requests |\n\nNote\n\nPremium requests for Spark and Copilot cloud agent are tracked in dedicated SKUs from November 1, 2025. This provides better cost visibility and budget control for each AI product.\n\nTip\n\nFor instructions on viewing how many premium requests you have used and advice on how to optimize usage, see [Monitoring your GitHub Copilot usage and entitlements (legacy)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/monitor-premium-requests).\n\n## [How do request allowances work per plan?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#how-do-request-allowances-work-per-plan)\n\nNote\n\nBilling for premium requests began on June 18, 2025, for all paid Copilot plans on GitHub.com, and on August 1, 2025, on GHE.com. The request counters were only set to zero for paid plans.\n\nIf you're on a **paid plan**, you get unlimited inline suggestions. Rate limiting is in place to accommodate for high demand. See [Usage limits for GitHub Copilot](https://docs.github.com/en/copilot/concepts/usage-limits).\n\nPaid plans also receive a monthly allowance of premium requests, which can be used for advanced chat interactions, inline suggestions using premium models, and other premium features.\n\n|  | Copilot Pro | Copilot Pro+ |\n| --- | --- | --- |\n| Pricing | $10 USD per month<br>(free for some users) | $39 USD per month |\n| Premium requests | 300 per month | 1500 per month |\n| Purchase additional premium requests at $0.04/request |  |  |\n\n## [What happens to unused requests at the end of the month?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-happens-to-unused-requests-at-the-end-of-the-month)\n\nUnused requests for the previous month do not carry over to the following month. Premium request counters reset on the 1st of each month at 00:00:00 UTC. See [Monitoring your GitHub Copilot usage and entitlements (legacy)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/monitor-premium-requests).\n\n## [What if I run out of premium requests?](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#what-if-i-run-out-of-premium-requests)\n\nNote\n\nAdditional premium requests are not available to:\n\n- Users who subscribe, or have subscribed, to Copilot Pro or Copilot Pro+ through GitHub Mobile on iOS or Android.\n\nIf you use all of your premium requests, you can still use Copilot with one of the included models for the rest of the month. This is subject to change. Response times for the included models may vary during periods of high usage. Requests to the included models may be subject to rate limiting. See [Usage limits for GitHub Copilot](https://docs.github.com/en/copilot/concepts/usage-limits).\n\nIf you need more premium requests beyond your monthly allowance set a budget for additional premium requests or upgrade to a higher plan. See [Setting up budgets to control spending on metered products](https://docs.github.com/en/billing/how-tos/set-up-budgets).\n\n## [Model multipliers](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests\\#model-multipliers)\n\nEach model has a premium request multiplier, based on its complexity and resource usage. Your premium request allowance is deducted according to this multiplier.\n\nSee [Model multipliers for annual plans on request-based billing (legacy)](https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/model-multipliers-for-annual-plans).",
          "metadata": {
            "viewport": "width=device-width, initial-scale=1",
            "og:site_name": "GitHub Docs",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "twitter:description": "Learn about requests in Copilot, including premium requests, how they work, and how to manage your usage effectively.",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogUrl": "https://docs-internal.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
            "ogTitle": "Requests in GitHub Copilot (legacy) - GitHub Docs",
            "status": "200",
            "og:title": "Requests in GitHub Copilot (legacy) - GitHub Docs",
            "og:url": "https://docs-internal.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
            "page-document-type": "article",
            "path-product": "copilot",
            "og:type": "article",
            "language": "en",
            "twitter:card": "summary",
            "page-content-type": "reference",
            "path-article": "copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
            "description": "Learn about requests in Copilot, including premium requests, how they work, and how to manage your usage effectively.",
            "twitter:domain": "docs-internal.github.com",
            "path-language": "en",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "twitter:url": "https://docs-internal.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
            "twitter:title": "Requests in GitHub Copilot (legacy) - GitHub Docs",
            "ogSiteName": "GitHub Docs",
            "path-version": "free-pro-team@latest",
            "title": "Requests in GitHub Copilot (legacy) - GitHub Docs",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-dbfe8710fd17",
            "sourceURL": "https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
            "url": "https://docs.github.com/en/copilot/reference/copilot-billing/request-based-billing-legacy/copilot-requests",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "proxyUsed": "basic",
            "cacheState": "hit",
            "cachedAt": "2026-09-15T17:23:32.056Z",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/copilot",
          "title": "GitHub Copilot documentation",
          "description": "You can use GitHub Copilot to enhance your productivity and assist as you work on code. Copilot subscription. Learn about Copilot",
          "position": 8,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# GitHub Copilot documentation\n\nYou can use GitHub Copilot to enhance your productivity and assist as you work on code.\n\n[Overview](https://docs.github.com/copilot/get-started/what-is-github-copilot) [Quickstart](https://docs.github.com/copilot/get-started/quickstart)\n\n## Articles\n\nCategory:All categories\n\n### [About agent apps](https://docs.github.com/en/copilot/concepts/agents/agent-apps)\n\nAgent apps let you use partner-built agents directly in your workflows on GitHub, powered by your Copilot subscription.\n\nLearn about CopilotAuthor and optimize with Copilot\n\n### [About agent management](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/agent-management)\n\nUse one centralized control page to jump between agent sessions, check progress, and stay in control without losing your place.\n\nLearn about Copilot\n\n### [About agent skills](https://docs.github.com/en/copilot/concepts/agents/about-agent-skills)\n\nSkills allow Copilot to perform specialized tasks.\n\nLearn about Copilot\n\n### [About cloud and local sandboxes for GitHub Copilot](https://docs.github.com/en/copilot/concepts/about-cloud-and-local-sandboxes)\n\nCloud and local sandboxes provide isolated execution environments that let Copilot safely interact with code, tools, filesystem, and network resources securely on your local machine or in fully isolated cloud environments.\n\nLearn about Copilot\n\n### [About Copilot auto model selection](https://docs.github.com/en/copilot/concepts/models/auto-model-selection)\n\nAutomatically select the best model for each task.\n\nLearn about Copilot\n\n### [About Copilot automations](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-automations)\n\nAutomations let you run Copilot cloud agent automatically, on a schedule or in response to events in a repository.\n\nLearn about CopilotIntegrate Copilot with your tools\n\n### [About Copilot integrations](https://docs.github.com/en/copilot/concepts/tools/about-copilot-integrations)\n\nIntegrate Copilot with other tools and platforms to streamline your workflow.\n\nLearn about Copilot\n\n### [About custom agents](https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-custom-agents)\n\nCustom agents enhance Copilot with assistance tailored to your needs.\n\nLearn about Copilot\n\n### [About custom agents](https://docs.github.com/en/copilot/concepts/agents/copilot-cli/about-custom-agents)\n\nCustom agents enhance Copilot with assistance tailored to your needs.\n\nShowing 1-9 of 438",
          "metadata": {
            "og:title": "GitHub Copilot documentation - GitHub Docs",
            "og:url": "https://docs-internal.github.com/en/copilot",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "ogSiteName": "GitHub Docs",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "viewport": "width=device-width, initial-scale=1",
            "path-article": "copilot",
            "language": "en",
            "path-language": "en",
            "page-document-type": "product",
            "twitter:domain": "docs-internal.github.com",
            "twitter:url": "https://docs-internal.github.com/en/copilot",
            "ogTitle": "GitHub Copilot documentation - GitHub Docs",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-product": "copilot",
            "status": "200",
            "page-content-type": "landing",
            "og:type": "article",
            "og:site_name": "GitHub Docs",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "twitter:card": "summary",
            "twitter:description": "You can use GitHub Copilot to enhance your productivity and assist as you work on code.",
            "ogUrl": "https://docs-internal.github.com/en/copilot",
            "title": "GitHub Copilot documentation - GitHub Docs",
            "description": "You can use GitHub Copilot to enhance your productivity and assist as you work on code.",
            "twitter:title": "GitHub Copilot documentation - GitHub Docs",
            "path-version": "free-pro-team@latest",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-de5ba216d1e0",
            "sourceURL": "https://docs.github.com/copilot",
            "url": "https://docs.github.com/en/copilot",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "proxyUsed": "basic",
            "cacheState": "hit",
            "cachedAt": "2026-09-14T02:47:13.196Z",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
          "title": "Quickstart for using GitHub Copilot in your IDE",
          "description": "Use Copilot in your IDE to explain concepts, complete code, propose edits, and validate files with agent mode.",
          "position": 9,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\nIn this article\n\n# Quickstart for using GitHub Copilot in your IDE\n\nUse Copilot in your IDE to explain concepts, complete code, propose edits, and validate files with agent mode.\n\nCopy markdown\n\n## In this article\n\nGitHub Copilot is available in supported IDEs: Visual Studio Code, Visual Studio, JetBrains IDEs, Eclipse, and XCode. It can answer questions and explain concepts, and suggest or complete code as you work. In IDEs that support agent mode, it can also propose edits and validate files.\n\nThis quickstart shows you how to use these capabilities. It takes about ten minutes.\n\n## [Prerequisites](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#prerequisites)\n\n- **An active Copilot plan.** See [About GitHub Copilot](https://docs.github.com/en/copilot/get-started/what-is-github-copilot#get-access).\n- **The Copilot extension for your IDE.** See [Installing the GitHub Copilot extension in your environment](https://docs.github.com/en/copilot/how-tos/set-up/install-copilot-extension).\n- **Signed in to GitHub in your IDE.** If you have authentication problems, see [Troubleshooting common issues with GitHub Copilot](https://docs.github.com/en/copilot/how-tos/troubleshoot-copilot/troubleshoot-common-issues).\n\n## [Ask a question about your code](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#ask-a-question-about-your-code)\n\n1. Open an existing code file in your editor.\n\n2. Open the Copilot Chat window.\n\nThe way you do this depends on your IDE.\n   - In Visual Studio Code, press `Ctrl` + `Alt` + `I` (Windows/Linux) or `Control` + `Command` + `I` (macOS).\n   - In other IDEs, look for Copilot Chat in the menu bar or sidebar. See [Asking GitHub Copilot questions in your IDE](https://docs.github.com/en/copilot/how-tos/chat-with-copilot/chat-in-ide).\n3. Type `what does this file do`, then press `Enter`.\n\nCopilot's answer displays below your question.\n\n4. Select a line of code in the editor, then type `explain this line` in the chat window.\n\n\nCopilot can see the file you have open, so you can ask about your code instead of describing it first.\n\n## [Let Copilot work through a task](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#let-copilot-work-through-a-task)\n\nMost IDEs support **agent mode**. In agent mode, you describe what you want, and Copilot decides which files to change, makes the edits, and runs commands with your approval.\n\nTo check whether your IDE supports agent mode, see [Copilot feature matrix](https://docs.github.com/en/copilot/reference/copilot-feature-matrix?tool=ides).\n\n1. In the chat window, select **Agent**, then describe a task. For example, `Create a task manager web app with the ability to add, delete, and mark tasks as completed`.\n\n2. Press `Enter`.\n\nCopilot generates the files and code needed.\n\n3. Review the changes, then use your IDE's controls to accept them.\n\n\nWhen you're ready, save your changes, and commit when everything looks good.\n\n## [Get your first inline suggestion](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#get-your-first-inline-suggestion)\n\nGitHub Copilot can suggest code as you work and supports many programming languages. This example uses JavaScript, but you can follow along in your preferred language.\n\n1. In your IDE, create a new JavaScript ( _\\*.js_) file.\n\n2. In the JavaScript file, type the following function header.\n\n\n\n```javascript\nfunction calculateDaysBetweenDates(begin, end) {\n```\n\n\n\nCopilot will automatically suggest an entire function body in grayed text. The exact suggestion may vary.\n\n3. To accept the suggestion, press `Tab`.\n\n\nWhen you're ready, save your changes, and commit when everything looks good.\n\n## [Next steps](https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide\\#next-steps)\n\n- Customize Copilot in your IDE with custom instructions. See [Adding repository custom instructions for GitHub Copilot in your IDE](https://docs.github.com/en/copilot/how-tos/configure-custom-instructions-in-your-ide/add-repository-instructions-in-your-ide#further-reading).\n- Try the GitHub Copilot app. See [Getting started with the GitHub Copilot app](https://docs.github.com/en/copilot/get-started/quickstart-copilot-app).",
          "metadata": {
            "ogSiteName": "GitHub Docs",
            "ogUrl": "https://docs-internal.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
            "path-version": "free-pro-team@latest",
            "og:title": "Quickstart for using GitHub Copilot in your IDE - GitHub Docs",
            "ogTitle": "Quickstart for using GitHub Copilot in your IDE - GitHub Docs",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "page-document-type": "article",
            "path-language": "en",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "page-content-type": "get-started",
            "title": "Quickstart for using GitHub Copilot in your IDE - GitHub Docs",
            "status": "200",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-product": "copilot",
            "twitter:domain": "docs-internal.github.com",
            "twitter:url": "https://docs-internal.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
            "viewport": "width=device-width, initial-scale=1",
            "og:type": "article",
            "og:url": "https://docs-internal.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
            "description": "Use Copilot in your IDE to explain concepts, complete code, propose edits, and validate files with agent mode.",
            "og:site_name": "GitHub Docs",
            "twitter:title": "Quickstart for using GitHub Copilot in your IDE - GitHub Docs",
            "twitter:description": "Use Copilot in your IDE to explain concepts, complete code, propose edits, and validate files with agent mode.",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "twitter:card": "summary",
            "language": "en",
            "path-article": "copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-e197d3431ca8",
            "sourceURL": "https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
            "url": "https://docs.github.com/en/copilot/get-started/quickstart-for-using-github-copilot-in-your-ide",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "timezone": "America/New_York",
            "proxyUsed": "basic",
            "cacheState": "miss",
            "indexId": "3dc9e98e-b2d1-4a93-a37d-b99b48ad35db",
            "creditsUsed": 1
          }
        },
        {
          "url": "https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
          "title": "Use GitHub Copilot CLI",
          "description": "Understand the different ways you can use GitHub Copilot in your terminal. Using GitHub Copilot CLI. Learn how to use GitHub Copilot from the command line.",
          "position": 10,
          "category": "github",
          "markdown": "[Skip to main content](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli#main-content)\n\n[Skip to content](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli#main-content)\n\nCollapse sidebarExpand sidebar\n\nScroll breadcrumbs left\n\nScroll breadcrumbs right\n\n# Use GitHub Copilot CLI\n\nUnderstand the different ways you can use GitHub Copilot in your terminal.\n\n## [Using GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/overview)\n\nLearn how to use GitHub Copilot from the command line.\n\n## [Allowing and denying tool use](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/allowing-tools)\n\nControl which tools Copilot CLI can use to avoid unintended changes.\n\n## [Use voice input with Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/voice-input)\n\nSpeak your prompts to GitHub Copilot CLI instead of typing them, using the CLI's speech-to-text feature.\n\n## [Connecting GitHub Copilot CLI to VS Code](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/connecting-vs-code)\n\nConnect Copilot CLI to VS Code to share context, trust settings, and output.\n\n## [Delegating tasks to Copilot](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/delegate-tasks-to-cca)\n\nUse autopilot mode, or the `/delegate` slash command, to get Copilot to work autonomously on your behalf.\n\n## [Browsing issues, pull requests, and gists from GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/browse-issues-prs-gists)\n\nUse the tabs in an interactive Copilot CLI session to browse issues, pull requests and gists, without leaving the terminal.\n\n## [Asking a side question in GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/ask-a-side-question)\n\nDuring a CLI session, you can ask Copilot a question without adding the prompt, or the answer, to your conversation history.\n\n## [Rolling back changes made during a GitHub Copilot CLI session](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/roll-back-changes)\n\nRewind your Copilot CLI session to a previous prompt to undo changes in conversation history, and optionally restore files.\n\n## [Working with multiple GitHub Copilot CLI sessions](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/work-with-multiple-sessions)\n\nRun several Copilot CLI sessions at once and use the sidebar to quickly move between them.\n\n## [Invoking custom agents](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/invoke-custom-agents)\n\nUse custom agents, skills, and MCP servers in Copilot CLI to extend its capabilities.\n\n## [Steering agents in GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/steer-agents)\n\nGuide Copilot during task execution to keep work on track with your intent.\n\n## [Steering a GitHub Copilot CLI session from another device](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/steer-remotely)\n\nEnable remote control for a Copilot CLI session so you can monitor progress, respond to prompts, and continue working from GitHub.com or GitHub Mobile.\n\n## [Setting an AI credit session limit in GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/set-session-limit)\n\nLimit the amount of AI credits Copilot can spend on a session to control costs and keep tasks predictable.\n\n## [Requesting a code review with GitHub Copilot CLI](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/agentic-code-review)\n\nUse Copilot CLI to review your code changes directly from the terminal.\n\n## [Managing pull requests with the /pr command](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/manage-pull-requests)\n\nUse the `/pr` slash command to view, create, and fix pull requests directly from Copilot CLI.\n\n## [Speeding up task completion with the /fleet command](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/speed-up-task-completion)\n\nLearn how you can speed up the completion of a multi-step implementation plan by using the `/fleet` slash command.\n\n## [Using GitHub Copilot CLI session data](https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli/chronicle)\n\nResume previous Copilot CLI sessions, use the `/chronicle` slash command to get insights from your session history, and ask Copilot questions about your CLI usage.",
          "metadata": {
            "twitter:url": "https://docs-internal.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
            "page-document-type": "subcategory",
            "ogTitle": "Use GitHub Copilot CLI - GitHub Docs",
            "ogSiteName": "GitHub Docs",
            "page-content-type": "how-tos",
            "twitter:domain": "docs-internal.github.com",
            "path-language": "en",
            "twitter:description": "Understand the different ways you can use GitHub Copilot in your terminal.",
            "og:type": "article",
            "ogUrl": "https://docs-internal.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
            "path-article": "copilot/how-tos/copilot-cli/use-copilot-cli",
            "docs-team-metrics": "copilot-cli",
            "language": "en",
            "twitter:card": "summary",
            "twitter:title": "Use GitHub Copilot CLI - GitHub Docs",
            "ogImage": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "viewport": "width=device-width, initial-scale=1",
            "og:url": "https://docs-internal.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
            "og:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "path-version": "free-pro-team@latest",
            "path-product": "copilot",
            "title": "Use GitHub Copilot CLI - GitHub Docs",
            "og:site_name": "GitHub Docs",
            "google-site-verification": "c1kuD-K2HIVF635lypcsWPoD4kilo5-jA_wBFyT4uMY",
            "twitter:image": "https://docs.github.com/assets/cb-345/images/social-cards/copilot.png",
            "og:title": "Use GitHub Copilot CLI - GitHub Docs",
            "description": "Understand the different ways you can use GitHub Copilot in your terminal.",
            "status": "200",
            "favicon": "https://docs.github.com/assets/cb-345/images/site/favicon.png",
            "scrapeId": "01a0a963-2711-75cd-83bd-e7947f48c744",
            "sourceURL": "https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
            "url": "https://docs.github.com/en/copilot/how-tos/copilot-cli/use-copilot-cli",
            "statusCode": 200,
            "contentType": "text/html; charset=utf-8",
            "timezone": "America/New_York",
            "proxyUsed": "basic",
            "cacheState": "miss",
            "indexId": "facae177-3323-47d6-885b-54ffdab612ec",
            "creditsUsed": 1
          }
        }
      ]
    },
    "creditsUsed": 12,
    "id": "01a0a963-1add-77ed-a51c-f15c91f9747a"
  }
}