The dependency graph is a summary of the manifest and lock files stored in a repository and any dependencies that are submitted for the repository using the Dependency submission API (beta). For each repository, it shows:
- Dependencies, the ecosystems and packages it depends on
- Dependents, the repositories and packages that depend on it
When you push a commit to GitHub Enterprise Cloud that changes or adds a supported manifest or lock file to the default branch, the dependency graph is automatically updated. In addition, the graph is updated when anyone pushes a change to the repository of one of your dependencies. For information on the supported ecosystems and manifest files, see "Supported package ecosystems" below.
Additionally, you can use the Dependency submission API (beta) to submit dependencies from the package manager or ecosystem of your choice, even if the ecosystem is not supported by dependency graph for manifest or lock file analysis. The dependency graph will display the submitted dependencies grouped by ecosystem, but separately from the dependencies parsed from manifest or lock files. For more information on the Dependency submission API, see "Using the Dependency submission API."
When you create a pull request containing changes to dependencies that targets the default branch, GitHub uses the dependency graph to add dependency reviews to the pull request. These indicate whether the dependencies contain vulnerabilities and, if so, the version of the dependency in which the vulnerability was fixed. For more information, see "About dependency review."
The dependency graph is available for every public repository that defines dependencies in a supported package ecosystem using a supported file format. Repository administrators can also set up the dependency graph for private repositories. For more information , see "Configuring the dependency graph."
You may not be able to enable or disable the dependency graph if an enterprise owner has set a policy at the enterprise level. For more information, see "Enforcing policies for code security and analysis for your enterprise."
The dependency graph includes all the dependencies of a repository that are detailed in the manifest and lock files, or their equivalent, for supported ecosystems, as well as any dependencies that are submitted using the Dependency submission API (beta). This includes:
- Direct dependencies, that are explicitly defined in a manifest or lock file or have been submitted using the Dependency submission API (beta)
- Indirect dependencies of these direct dependencies, also known as transitive dependencies or sub-dependencies
The dependency graph identifies indirect dependencies either explicitly from a lock file or by checking the dependencies of your direct dependencies. For the most reliable graph, you should use lock files (or their equivalent) because they define exactly which versions of the direct and indirect dependencies you currently use. If you use lock files, you also ensure that all contributors to the repository are using the same versions, which will make it easier for you to test and debug code.
For more information on how GitHub Enterprise Cloud helps you understand the dependencies in your environment, see "About supply chain security."
For public repositories, only public repositories that depend on it or on packages that it publishes are reported. This information is not reported for private repositories.
You can use the dependency graph to:
- Explore the repositories your code depends on, and those that depend on it. For more information, see "Exploring the dependencies of a repository."
- View a summary of the dependencies used in your organization's repositories in a single dashboard. For more information, see "Viewing insights for your organization."
- View and update vulnerable dependencies for your repository. For more information, see "About Dependabot alerts."
- See information about vulnerable dependencies in pull requests. For more information, see "Reviewing dependency changes in a pull request."
The recommended formats explicitly define which versions are used for all direct and all indirect dependencies. If you use these formats, your dependency graph is more accurate. It also reflects the current build set up and enables the dependency graph to report vulnerabilities in both direct and indirect dependencies. Indirect dependencies that are inferred from a manifest file (or equivalent) are excluded from the checks for insecure dependencies.
|Package manager||Languages||Recommended formats||All supported formats|
|NuGet||.NET languages (C#, F#, VB), C++|
|GitHub Actions workflows[†]||YAML|
[†] GitHub Actions workflows must be located in the
.github/workflows/ directory of a repository to be recognized as manifests. Any actions or workflows referenced using the syntax
jobs.<job_id>.uses will be parsed as dependencies. For more information, see "Workflow syntax for GitHub Actions."
[‡] If you list your Python dependencies within a
setup.py file, we may not be able to parse and list every dependency in your project.
Note: GitHub Actions workflow dependencies are displayed in the dependency graph for informational purposes. Dependabot alerts are not currently supported for GitHub Actions workflows.
You can use the Dependency submission API (beta) to add dependencies from the package manager or ecosystem of your choice to the dependency graph, even if the ecosystem is not in the supported ecosystem list above. The dependency graph will display the submitted dependencies grouped by ecosystem, but separately from the dependencies parsed from manifest or lock files. You will only get Dependabot alerts for dependencies that are from one of the supported ecosystems of the GitHub Advisory Database. For more information on the Dependency submission API, see "Using the Dependency submission API."