#082Tutorial3 min

Integrate Hackaiz into GitHub Actions

Choose a credit-using component gate or a saved-platform workflow, protect secrets and interpret the result correctly.

Hackaiz team · updated 21/11/2026

Illustration: a developer uses an editor and a terminal.
Illustration: a developer uses an editor and a terminal. Illustrative photo, not a screenshot.

01Choose what the workflow is meant to verify

The integrations page describes the Hackaiz Import Gate action as Hacktualiz/import-gate@v1, using a personal key to check submitted components. It also provides a different connected template at https://hackaiz.com/hackaiz-connected-action.yml. That template requests an analysis of a platform already saved in Hackaiz. These are not interchangeable tests: component checking concerns submitted text, while connected analysis concerns the saved authorized target.

The project also contains a separate downloadable workflow that applies open-source secret and code rules. Do not use that local-rule workflow’s behavior to explain a credit-using account analysis. Decide which mechanism you are configuring and name it clearly in your repository. In particular, do not label a successful saved-platform run as verification of the pull request commit. The connected template explicitly states that it does not verify the pull request head commit.

02Prepare the component-check action deliberately

The documented setup adds HACKAIZ_API_KEY in GitHub Settings, Secrets and variables, Actions, then creates .github/workflows/hackaiz.yml. The action step uses “uses: Hacktualiz/import-gate@v1” with the input “api-key: ${{ secrets.HACKAIZ_API_KEY }}”. GitHub’s Choose a workflow page does not list every action as a ready-made workflow, so create a workflow file rather than assuming a missing template means the integration does not exist.

The integrations page says Danger or an incomplete result fails the action by default. Each paid component check uses credits. SARIF output can be uploaded to GitHub’s Security interface when repository permissions and code-scanning options allow it; the action result alone does not guarantee that interface is available. Static component checking does not execute the submitted code and cannot establish the runtime behavior of all installed dependencies.

03Understand the saved-platform template before copying it

The connected template expects the prepared CLI source at tools/hackaiz-cli. The integration page says that client is not published on npm and autonomous processing must be deployed and synchronized; a real GitHub execution is not confirmed there. The template uses HACKAIZ_API_KEY as a secret and HACKAIZ_PROJECT_ID as a repository variable. It builds a stable request key from repository, pull request and commit information, but that key does not change the target into a commit-specific scan.

The workflow checks out the default branch for the prepared client, excludes pull requests from forks, launches the saved platform and can comment a report link. It does not modify source or merge changes. Its launch step does not run the CLI verify command. Because analyze can finish with a partial report, a green job must not be presented as proof that a complete security verification passed. Add a separately reviewed verification design if your release policy needs one.

04Protect credentials and keep the release decision human

Never expose the personal key to untrusted fork code. Keep the fork exclusion and the separation between trusted automation and proposed source changes. Use the minimum permissions needed for the selected workflow; the connected template requests pull-request write permission to post a comment, not permission to rewrite the application. Keep keys out of logs and artifacts. Review any workflow changes before allowing them to run with account access.

A website or API target still requires domain proof before analysis through DNS TXT _hackaiz, the HTTPS verification file or the project AI creating that file. GitHub repository authorization does not replace proof. Read completion, coverage and findings before approving release. If a finding leads to a correction, Hackaiz creates a draft GitHub pull request for human review and approval, never an automatic merge. A workflow is one review input, not certification or a guarantee that the application is secure.

Frequently asked questions

Was this article helpful?

Related articles