@adguard/github-actions-runner
is a versatile command-line interface (CLI) tool designed to facilitate automation with
GitHub Actions. It simplifies tasks such as repository mirroring, command execution, and artifact management. By
integrating with GitHub's robust API, it provides a seamless approach to running and managing workflows, ideal for
developers and DevOps engineers looking to enhance their CI/CD processes.
This tool is designed to offer greater flexibility when interacting with GitHub Actions, allowing you to automate
workflows, manage artifacts, and pass secrets securely. Whether you're a DevOps engineer or a software
developer, @adguard/github-actions-runner
enables you to focus on coding and development, reducing the overhead of
managing pipelines.
To get started with using @adguard/github-actions-runner
, ensure that Node.js is installed on your system, then
install the package with one of the following methods:
- Using pnpm:
pnpm install @adguard/github-actions-runner
- Using npm:
npm install @adguard/github-actions-runner
- Using yarn:
yarn add @adguard/github-actions-runner
The CLI tool can trigger and manage GitHub Actions with the following command format:
Usage: github-actions-runner run-action [options]
Triggers a GitHub Action workflow run
Options:
-r, --repo <repo> repository specified as "owner/repo", e.g., "AdguardTeam/GithubActionsRunner"
-w, --workflow <workflow> workflow file to trigger, e.g., "test.yml"
-b, --branch <branch> branch name
-c, --rev <revision> commit revision
-a, --artifacts-path [artifacts-path] local path for downloading artifacts; if not specified, artifacts will not be downloaded
--commit-timeout [timeout-sec] timeout in seconds to wait for the commit to appear in the repository (default: 300 seconds)
--branch-timeout [timeout-sec] timeout in seconds to wait for the branch to appear in the repository (default: 300 seconds)
--workflow-run-creation-timeout [timeout-sec] timeout in seconds to wait for the workflow run to be created (default: 300 seconds)
--workflow-run-completion-timeout [timeout-sec] timeout in seconds to wait for the workflow run to be completed (default: 300 seconds)
-s, --secret <KEY=VALUE> Secret key-value pair for the GitHub Action workflow, e.g., "API_KEY=12345".You can add multiple secrets by repeating the option. (default: [])
--sync-secrets Sync secrets with the repository secrets. Secrets which were not provided will be removed (default: false)
-v, --verbose enable verbose mode (default: false)
-h, --help display help for command
Ensure that you have a valid GITHUB_TOKEN
set in your environment variables for authentication.
To run a GitHub Action workflow with the specified configuration:
github-actions-runner run-action \
--repo "AdguardTeam/GithubActionsRunner" \
--workflow "build.yml" \
--branch "master" \
--rev "abc1234" \
--artifacts-path "./downloads" \
--commit-timeout 300 \
--branch-timeout 300 \
--workflow-run-completion-timeout 300 \
--secret "API_KEY=12345" \
--secret "API_KEY2=67890" \
--verbose
This CLI tool is available as a Docker image, enabling its use in containerized environments. There are two Docker images available, each serving different use cases.
The standard Docker image includes an ENTRYPOINT
to simplify command execution. You can pull it from the GitHub
Container Registry and run it like this:
docker pull ghcr.io/adguard/github-actions-runner:latest
To run the image, use the following command:
docker run --rm -it ghcr.io/adguard/github-actions-runner:latest run-action \
--repo "AdguardTeam/GithubActionsRunner" \
--workflow "build.yml" \
--branch "master" \
--rev "abc1234" \
--artifacts-path "./downloads" \
--commit-timeout 300 \
--branch-timeout 300 \
--verbose
This alternative Docker image doesn't set an ENTRYPOINT
, providing more flexibility for advanced use cases. Pull it
from the GitHub Container Registry and run it like this:
docker pull ghcr.io/adguard/github-actions-runner:latest-ci
To run the image, you need to specify the command manually:
docker run --rm -it ghcr.io/adguard/github-actions-runner:latest-ci /bin/sh
In this mode, you can execute commands manually inside the container or create custom entry points for specific workflows.
Contributions to extend the functionality or improve the tool are welcome. Please refer to the project's repository on GitHub to submit issues or pull requests.