About secret scanning
Secret scanning is a security feature that helps detect and prevent the accidental inclusion of sensitive information such as API keys, passwords, tokens, and other secrets in your repository. When enabled, secret scanning scans commits in repositories for known types of secrets and alerts repository administrators upon detection.
Secret scanning scans your entire Git history on all branches present in your GitHub repository for secrets, even if the repository is archived. GitHub will also periodically run a full Git history scan for new secret types in existing content in public repositories where secret scanning is enabled when new supported secret types are added.
Additionally, secret scanning scans:
- Descriptions et commentaires dans les problèmes
- Titres, descriptions et commentaires, dans les problèmes historiques ouverts et fermés. Une notification est envoyée au partenaire approprié quand un modèle de partenaire historique est détecté.
- Titres, descriptions et commentaires dans les demandes de tirage
- Titres, descriptions et commentaires dans GitHub Discussions
- Wikis
- Idées secrètes. Une notification est envoyée au partenaire concerné lorsqu’un modèle partenaire est détecté dans un gist secret.
This additional scanning is free for public repositories.
Conseil
Regardless of the enablement status of Advanced Security features, organizations on GitHub Team and GitHub Enterprise can run a free report to scan the code in the organization for leaked secrets.
To generate a report, open l'onglet ** Sécurité ** de votre organisation, affichez la page ** Évaluations **, puis cliquez sur Analyser votre organisation.
When a supported secret is leaked, GitHub generates a secret scanning alert. Alerts are reported on the Security tab of repositories on GitHub, where you can view, evaluate, and resolve them. For more information, see Gestion des alertes à partir de l’analyse des secrets.
Service providers can partner with GitHub to provide their secret formats for scanning. We automatically run secret scanning for partner patterns on all public repositories and public npm packages. Pour en savoir plus sur notre programme de partenariat, consultez Programme de partenariat d’analyse des secrets.
Any strings that match patterns that were provided by secret scanning partners are reported directly to the relevant partner, and aren't displayed on GitHub. For more information about partner patterns, see About secret scanning alerts.
For information about the secrets and service providers supported by secret scanning, see Modèles de détection de secrets pris en charge.
You can use the REST API to monitor results from secret scanning across your repositories or organization. For more information about API endpoints, see Points de terminaison d’API REST pour l’analyse de secrets.
You can also use security overview to see an organization-level view of which repositories have enabled secret scanning and the alerts found. For more information, see About security overview.
Vous pouvez auditer les actions effectuées en réponse aux alertes secret scanning à l’aide des outils GitHub. Pour plus d’informations, consultez « Auditing security alerts ».
How secret scanning works
Below is a typical workflow that explains how secret scanning works:
-
Detection: Secret scanning automatically scans your repository's contents for sensitive data, such as API keys, passwords, tokens, and other secrets. It looks for patterns and heuristics that match known types of secrets.
-
Alerts: When a potential secret is detected, GitHub generates an alert and notifies the relevant repository administrators and users. This notification includes details about the detected secret, such as its location in the repository. For more information about alert types and alert details, see About secret scanning alerts.
-
Review: When a secret is detected, you'll need to review the alert details provided.
-
Remediation: You then need to take appropriate action to remediate the exposure. This should always include rotating the affected credential to ensure it is no longer usable. It may also include removing the secret from the repository's history (using tools like
git-filter-repo; see Suppression de données sensibles dans un dépôt for more details) though this will likely involve a heavy cost in time and effort, and is usually unnecessary if the credentials have been revoked. -
Monitoring: It's good practice to regularly audit and monitor your repositories to ensure no other secrets are exposed.
-
Integration with partners: GitHub works with various service providers to validate secrets. When a partner secret is detected, GitHub notifies the provider so they can take appropriate action, such as revoking the credential. For more information about the partnership program, see Programme de partenariat d’analyse des secrets.
About the benefits of secret scanning
-
Enhanced security: Secret scanning scans your repositories for sensitive information like API keys, passwords, tokens, and other secrets. By detecting these early, you can mitigate potential security risks before they are exploited by malicious actors.
-
Automated detection: The feature automatically scans your codebase, including commits, issues, and pull requests, ensuring continuous protection without requiring manual intervention. This automation helps in maintaining security even as your repository evolves.
-
Real-time alerts: When a secret is detected, secret scanning provides real-time alerts to repository administrators and contributors. This immediate feedback allows for swift remediation actions.
-
Integration with service providers: GitHub partners with various service providers to validate detected secrets. When a secret is identified, GitHub notifies the corresponding service provider to take appropriate actions, such as revoking the exposed credential. For more information, see Programme de partenariat d’analyse des secrets.
-
Custom pattern support: Organizations can define custom patterns to detect proprietary or unique types of secrets that may not be covered by default patterns. This flexibility allows for tailored security measures specific to your environment.
-
Ability to detect non-provider patterns: You can expand the detection to include non-provider patterns such as connection strings, authentication headers, and private keys, for your repository or organization.
Customizing secret scanning
Once secret scanning is enabled, you can customize it further:
Detection of non-provider patterns
Scan for and detect secrets that are not specific to a service provider, such as private keys and generic API keys. For more information, see Activation de l’analyse du secret des modèles non fournisseurs.
Performing validity checks
Validity checks help you prioritize alerts by telling you which secrets are active or inactive. For more information, see Activer les vérifications de validité pour votre référentiel and Évaluation des alertes à partir de l’analyse des secrets.
Defining custom patterns
Define your own patterns for secrets used by your organization that secret scanning can scan for and detect. For more information, see Définition de modèles personnalisés pour l’analyse des secrets.
Analyse des secrets Copilot
- Détection de secrets génériques: Leverage secret scanning's AI capabilities to detect unstructured secrets, such as passwords, in your repository. For more information, see Détection responsable des secrets génériques avec l’analyse des secrets Copilot.
- Générateur d’expressions régulières: Leverage secret scanning's AI capabilities to generate regular expressions that will capture all your custom patterns. For more information, see Génération responsable d’expressions régulières avec l’analyse des secrets Copilot.
Further reading
- Activation de l’analyse des secrets pour votre référentiel
- About push protection
- Travailler avec l’analyse des secrets et la protection push
- Best practices for preventing data leaks in your organization
- Guide de démarrage rapide pour sécuriser votre référentiel
- Maintenir votre compte et vos données sécurisés