Red Hat Dependency Analytics (RHDA) Github Actions gives you awareness to security concerns when doing code commits to your GitHub repository. The RHDA platform uses vulnerability data sources to report the most up-to-date vulnerability information available.
Dependency Analytics uploads a report to the GitHub repository as an artifact, and as a SARIF file. Repository maintainers can find discovered vulnerabilities on the Security tab.
Ecosystem | Required Binaries and Prerequisites | Supported Manifests / Files |
---|---|---|
Java - Maven | mvn |
pom.xml |
Gradle - Gradle Installation | gradle |
build.gradle |
JavaScript - Npm | npm |
package.json |
Golang - Go Modules | go |
go.mod |
Python - pip Installer | pip |
requirements.txt |
docker | syft ,skopeo , Java version 20 or later |
Dockerfile |
Podman | syft ,skopeo , Java version 20 or later |
Containerfile |
In future releases, Red Hat plans to support other programming languages.
You can configure GitHub Actions to use RHDA for your coding environment. You can see examples of scanning for each supported language, or you can use the example below.
- GitHub Actions for your environment:
- Go
- Java
- Node.js
- Python
- For
syft
you can use the following script:- name: Setup syft run: | curl -sSfL https://raw.githubusercontent.com/anchore/syft/main/install.sh | sh -s -- -b /usr/local/bin
- For
skopeo
you can use the following script:- name: Setup skopeo run: | sudo apt-get -y update sudo apt-get -y install skopeo
Procedure
- Add the RHDA scanner to your project's GitHub Actions configuration:
- name: RHDA Scan id: rhda_scan uses: RHEcosystemAppEng/rhda-github-action@main
Example
This Node.js example shows you how to use the RHDA scanner, and uploading the results to GitHub.
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Setup Node.js
uses: actions/setup-node@v4
with:
node-version: 20
- name: RHDA Scan
id: rhda_scan
uses: RHEcosystemAppEng/rhda-github-action@main
An example screenshot of a Node.js RHDA scan:
Input | Description | Default |
---|---|---|
manifest_directory |
Path to the directory containing the manifest_file . |
|
manifest_file |
Name (basename ) of the manifest file to analyze. This file must exist in the manifest_directory . |
|
rhda_report_name |
The saved file name of the Red Hat Dependency Analytics report. | redhat-dependency-analytics-report |
github_token |
GitHub token to upload a SARIF file to the GitHub. | ${{ github.token }} |
upload_sarif |
Upload the generated SARIF file. The default is true . If you do not want to upload a SARIF file, then set the input to false . |
true |
upload_artifact |
Upload the generated RHDA report JSON file and SARIF file as an artifact. | true |
artifact_filename |
File name of the artifact to upload. The default name is, redhat-dependency-analytics-report . |
redhat-dependency-analytics-report |
fail_on |
A found vulnerability fails the workflow within the project. To fail the workflow when the vulnerability severity level is either error or warning , set this to warning . The default severity level is error . If you do not want to fail the action set this input to never . |
error |
use_python_virtual_environment |
Automates the installation of missing packages in a Python virtual environment when set to true. | false |
use_go_mvs |
Use the Minimal Version Selection (MVS) algorithm to select a set of module versions to use when building Go packages. | false |
enable_python_best_efforts_installation |
Installs Python packages tailored to the Python version in use, disregarding declared versions. This requires setting the Match Manifest Versions option to false , and the Use Python Virtual Environment option to true . |
false |
use_pip_dep_tree |
Use lightweight pipdeptree command line tool as the data source for building the Python dependency tree. This might significantly enhance analysis time. |
false |
match_manifest_version |
Restricts RHDA from performing analysis on dependency tags that do not match the tags requested within the manifest files. | false |
mvn_executable_path |
Specifies absolute path of mvn binary. |
mvn |
gradle_executable_path |
Specifies absolute path of gradle binary. |
gradle |
npm_executable_path |
Specifies absolute path of npm binary. |
npm |
go_executable_path |
Specifies absolute path of go binary. |
go |
python3_executable_path |
Specifies absolute path of python3 binary, python3 takes precedence over python . |
python3 |
pip3_executable_path |
Specifies absolute path of pip3 binary, pip3 takes precedence over pip . |
pip3 |
python_executable_path |
Specifies absolute path of python binary, python3 takes precedence over python . |
python |
pip_executable_path |
Specifies absolute path of pip binary, pip3 takes precedence over pip . |
pip |
syft_executable_path |
Specifies absolute path of syft binary. |
syft |
syft_config_path |
Specifies absolute path to the Syft configuration file. | |
skopeo_executable_path |
Specifies absolute path of skopeo binary. |
skopeo |
skopeo_config_path |
Specifies absolute path to the authentication file used by skopeo inspect . |
|
docker_executable_path |
Specifies absolute path of docker binary. |
docker |
podman_executable_path |
Specifies absolute path of podman binary. |
podman |
image_platform |
Specifies platform used for multi-arch images. |
- rhda_report_json: Path to the generated Red Hat Dependency Analytics report in JSON format.
- rhda_report_sarif: Path to the generated Red Hat Dependency Analytics report in SARIF format.
- artifact_id: The identifier of the uploaded artifact.
This action can also run RHDA scans on pull requests.
Because the action must check out the pull request's code to scan it, you must use a pull_request_target
trigger.
After the workflow runs, and labels the scanning results to the pull request.
The following screenshot shows vulnerability details in the GitHub UI for a pull request.
Use the following snippet to enable pull request scans in your repository:
on:
pull_request_target:
# These types are all required for RHDA to scan pull requests correctly and securely.
types: [ opened, synchronize, reopened, labeled, edited ]