About push protection
Push protection is a secret scanning feature that is designed to prevent sensitive information, such as secrets or tokens, from being pushed to your repository in the first place. Unlike secret scanning, which detects secrets after they have been committed, push protection proactively scans your code for secrets during the push process and blocks the push if any are detected.
Push protection helps you avoid the risks associated with exposed secrets, like unauthorized access to resources or services. With this feature, developers get immediate feedback and can address potential issues before they become a security concern.
You can enable push protection:
- At repository/organization level, if you are a repository administrator or an organization owner. You will see alerts in the Security tab of your repository when a contributor to the repository bypasses push protection.
- For your account on GitHub, as a user. This type of push protection is referred to as "push protection for users." It protects you from pushing secrets to any public repository on GitHub, but no alerts are generated.
ヒント
Regardless of the enablement status of push protection, organizations on GitHub Team and GitHub Enterprise can run a free report to scan the code in the organization for leaked secrets. The report also tells you how many secret leaks in your organization could have been prevented by push protection. See About secret security with GitHub.
For information about the secrets and service providers supported by push protection, see サポートされているシークレット スキャン パターン.
Push protection has some limitations. For more information, see シークレット スキャンのトラブルシューティング.
How push protection works
Push protection blocks secrets detected in:
- Pushes from the command line. See コマンド ラインからのプッシュ保護を操作する.
- Commits made in the GitHub UI. See GitHub UI でのプッシュ保護の操作.
- File uploads to a repository on GitHub.
- Requests to the REST API. See Working with push protection from the REST API.
- Interactions with the GitHub MCP server (public repositories only). See Working with push protection and the GitHub MCP server.
Once enabled, if push protection detects a potential secret during a push attempt, it will block the push and provide a detailed message explaining the reason for the block. You will need to review the code in question, remove any sensitive information, and reattempt the push.
By default, anyone with write access to the repository can choose to bypass push protection by specifying one of the bypass reasons outlined in the table. 共同作成者がシークレットのプッシュ保護ブロックをバイパスする場合、GitHub では次のことが行われます。
- リポジトリの [セキュリティ] タブでアラートを作成する。
- 監査ログにバイパス イベントを追加します。
- リポジトリを監視している organization または個人用アカウントの所有者、セキュリティ マネージャー、リポジトリ管理者に、シークレットへのリンクとそれが許可された理由を含むメール アラートを送信する。
次の表は、ユーザーがプッシュ保護ブロックをバイパスする方法ごとのアラートの動作を示しています。
| バイパスの理由 | アラート動作 |
|---|---|
| It's used in tests (テストで使用) | GitHub は、"テストで使用" として解決されたクローズしたアラートを作成します |
| It's a false positive (偽陽性) | GitHub は、"偽陽性" として解決されたクローズしたアラートを作成します |
| I'll fix it later (後で修正) | GitHub は、オープンなアラートを作成します |
If you want greater control over which contributors can bypass push protection and which pushes containing secrets should be allowed, you can enable delegated bypass for push protection. Delegated bypass lets you configure a designated group of reviewers to oversee and manage requests to bypass push protection from contributors pushing to the repository. For more information, see About delegated bypass for push protection.
You can also bypass push protection using the REST API. For more information, see シークレット スキャン用の REST API エンドポイント.
About the benefits of push protection
-
Preventative security: Push protection acts as a frontline defense mechanism by scanning code for secrets at the time of the push. This preventative approach helps to catch potential issues before they are merged into your repository.
-
Immediate feedback: Developers receive instant feedback if a potential secret is detected during a push attempt. This immediate notification allows for quick remediation, reducing the likelihood of sensitive information being exposed.
-
Reduced risk of data leaks: By blocking commits that contain sensitive information, push protection significantly reduces the risk of accidental data leaks. This helps in safeguarding against unauthorized access to your infrastructure, services, and data.
-
Efficient secret management: Instead of retrospectively dealing with exposed secrets, developers can address issues at the source. This makes secret management more efficient and less time-consuming.
-
Ability to detect custom patterns: Organizations can define custom patterns for detecting secrets unique to their environment. This customization ensures that push Protection can effectively identify and block even non-standard secrets.
-
Delegated bypass for flexibility: For cases where false positives occur or when certain patterns are necessary, the delegated bypass feature allows designated users to approve specific pushes. This provides flexibility without compromising overall security.
Every user across GitHub can also enable push protection for themselves within their individual settings. Enabling push protection for your user account means that your pushes are protected whenever you push to a public repository on GitHub, without relying on that repository to have push protection enabled. For more information, see ユーザーのプッシュ保護.
Customizing push protection
Once push protection is enabled, you can customize it further:
Configure push protected patterns
Customize which secret patterns are included in push protection at the enterprise or organization level. See Enterprise 用の追加のシークレット スキャン設定の構成 and 組織のグローバル セキュリティ設定の構成.
Define custom patterns
Define custom patterns that push protection can use to identify secrets and block pushes containing these secrets. For more information, see シークレット スキャンのカスタム パターンの定義.
Configure delegated bypass
Define contributors who can bypass push protection and add an approval process for other contributors. For more information, see About delegated bypass for push protection.