Skip to content

security

Subscribe to all “security” posts via RSS or follow GitHub Changelog on Twitter to stay updated on everything we ship.

→ ~ cd github-changelog
→ ~/github-changelog|main git log main
showing all changes successfully

Secret scanning’s push protection is available on public repositories, for free

Secret scanning's push protection feature is now generally available for all free public repositories on GitHub.com.

You can enable push protection for any public repository on GitHub.com from your repository's "Code security and analysis" settings in the UI or REST API. If you're an organization or enterprise owner, you can also also bulk-enable secret scanning.

For your repositories that are not a part of an organization, you can bulk-enable secret scanning and push protection in your personal "Code security and analysis" settings.

See more

Secret scanning’s push protection now generally available for GitHub Advanced Security

Secret scanning's push protection feature is now generally available for GitHub Advanced Security customers.

Customers can enable push protection for any private repository that has GitHub Advanced Security. Push protection can also be enabled for any public repository, for free. To bulk enable push protection, customers can visit their organization and enterprise's "Code security and analysis" settings in the UI or REST APIs.

Push protection is also available for any custom pattern defined at the repository, organization, and enterprise level. See step 11 under "Defining a custom pattern for a repository" for more details in our documentation.

See more

Dependabot alerts now automatically dismiss false positives for npm (public beta)

Starting today, Dependabot will be able to auto-dismiss npm alerts that have limited impact (e.g. long-running tests) or are unlikely to be exploitable. With this ship, Dependabot will cut false positives and reduce alert fatigue substantially.

On-by-default for public repositories, and opt-in for private repositories, this feature will result in 15% of low impact npm alerts being auto-dismissed moving forward – so you can focus on the alerts that matter, without worrying about the ones that don’t.

What’s changing?

When the feature is enabled, Dependabot will auto-dismiss certain types of vulnerabilities that are found in npm dependencies used in development (npm devDependency alerts with scope:development). This feature will help you proactively filter out false positives on development-scoped (non-production or runtime) alerts without compromising on high risk devDependency alerts.

Dependabot alerts auto-dismissal list view

Frequently asked questions

Why is GitHub making this change?

At GitHub, we’ve been thinking deeply about how to responsibly address long-running issues around alert fatigue and false positives. Rather than over-indexing on one criterion like reachability or dependency scope, we believe that a responsibly-designed solution should be able to detect and reason on a rich set of complex, contextual alert metadata.

That’s why, moving forward, we’re releasing a series of ships powered by an underlying, all-new, flexible and powerful alert rules engine. Today’s ship, our first application, leverages GitHub-curated vulnerability patterns to help proactively filter out false positive alerts.

Why auto-dismissal, rather than purely suppressing these alerts?

Auto-dismissing ensures any ignored alerts are 1) able to be reintroduced if alert metadata changes, 2) caught by existing reporting systems and workflows, and 3) extensible as a whole to future rules-based actions, where Dependabot can decision on subsets of alerts and do things like reopen for patch, open a Dependabot pull request, or even auto-merge if very risky.

How does GitHub identify and detect low impact alerts?

Auto-dismissed alerts match GitHub-curated vulnerability patterns. These patterns take into account contextual information about how you’re using the dependency and the level of risk they may pose to your repository. To learn more, see our documentation on covered classes of vulnerabilities.

How will this activity be reported?

Auto-dismissal activity is supported across webhooks, REST, GraphQL, and the audit log for Dependabot alerts. In addition, you can review your closed alert list with the resolution:auto-dismissed filter.

How will this experience look and feel?

Alerts identified as false positives will be automatically dismissed without a notification or new pull request, and appear as special timeline event. As these alerts are closed, you’ll still be able to review any auto-dismissed alerts with the resolution:auto-dismissed filter.

How do I reopen an automatically dismissed alert?

Like any manually dismissed alert, you can reopen an auto-dismissed alert from the alert list view or details page. This specific alert won’t be auto-dismissed again.

What happens if alert metadata changes or advisory information is withdrawn?

Dependabot recognizes and immediately responds to any changes to metadata which void auto-dismissal logic. For example, if you change the dependency scope and the alert no longer meets the criteria to be auto-dismissed, the alert will automatically reopen.

How can I enable or disable the feature?

This feature is on-by-default for public repositories and opt-in for private repositories. Repository admins can opt in or out from your Dependabot alerts settings in the Code Security page.

Is this feature available for enterprise?

Yes! In addition to all free repositories, this feature will ship immediately to GHEC and to GHES in version 3.10.

What’s next?

Next, we’ll expose our underlying engine – which enables Dependabot to perform actions based on a rich set of contextual alert metadata – so you can write your own custom rules to better manage your alerts, too.

How do I learn more?

How do I provide feedback?

Let us know what you think by providing feedback — we’re listening!

See more

npm provenance public beta

npm packages built on a cloud CI/CD system (like GitHub Actions) can now publish with provenance, meaning the package has verifiable links back to its source code and build instructions.

The cloud CI/CD system securely communicates this information by sending provenance information in a signed OIDC JWT to Sigstore's public-good servers, which returns a signing certificate that is sent to the registry along with your built package.

Here's an example of how to do a build with provenance in a GitHub Actions workflow:

name: Publish Package to npmjs
on:
 release:
   types: [created]
jobs:
 build:
   runs-on: ubuntu-latest
   permissions:
     contents: read
     id-token: write
   steps:
     - uses: actions/checkout@v3
     - uses: actions/setup-node@v3
       with:
         node-version: '18.x'
         registry-url: 'https://registry.npmjs.org'
     - run: npm install -g npm
     - run: npm ci
     - run: npm publish --provenance --access public
       env:
         NODE_AUTH_TOKEN: ${{ secrets.NPM_TOKEN }}

Once published, packages display provenance on the registry website:

Provenance displayed on the registry website

Dependencies with provenance can also be verified from the command line with npm audit signatures.

For more information, see generating provenance.

See more

Fixed bug that allowed a hovercard URL to be used to display the name, description, and star count of any repository

On March 30, 2023, we fixed a bug that allowed a dependency graph hovercard URL to be used to retrieve the name, description, and star count of any repository on GitHub.com. The bug was introduced on March 28, 2023 and our investigation has found no evidence of exploitation. To exploit the bug, a specific header needed to be set when making a request to the URL and the numeric ID of a repository provided. The URL would then return the HTML content designed to be used for a hovercard UI element with the repository name, description, and star count in the response.

This bug was reported to GitHub via the GitHub Bug Bounty program.

See more

Secret scanning now detects secrets leaked historically in issues

GitHub Advanced Security customers using secret scanning can now view any secrets exposed historically in an issue's title, description, or comments within the UI or the REST API. This expanded coverage will also detect and surface secrets matching any custom pattern defined at the repository, organization, or enterprise levels.

See more

Secret scanning now shows metadata for active GitHub tokens

Users with access to secret scanning alerts can now view metadata for any active GitHub token leaked in their repositories. Metadata includes details like the token's owner, expiration date, and access permissions. With this information, security teams can assess a leak's potential impact and prioritize remedial action accordingly.

This feature builds on our previous release in January, which introduced validity checks for leaked GitHub tokens.

github

See more

Secret scanning now shows metrics for custom patterns

GitHub Advanced Security users can now view alert metrics for custom patterns at the repository, organization, and enterprise levels directly from the custom pattern's page. Custom patterns with push protection enabled also show metrics like total secrets blocked and bypassed.

We welcome feedback in our code security discussion.

custom pattern metrics

See more

Repository security advisories REST API

You can now programmatically view and act on repository advisories via a new REST API. New endpoints to create, view, list, and update advisories are available to all. Additionally, new webhooks have been introduced that will alert maintainers when advisories are published or when a private vulnerability report is submitted.

Current advisory permissions extend to API usage.

See more

Organization APIs for fine-grained PATs management

Organization owners can now automate the approval and auditing of fine-grained personal access tokens (PATs) in their organization using a GitHub app. New APIs and webhook events allow a GitHub app to be notified of new PAT requests in an organization, review the request, and then approve or deny the PAT. They also provide a view of all approved fine-grained PATs for an organization, with the ability to revoke their authorization as well. These APIs and events are part of the ongoing fine-grained PAT public beta that launched last year.

Mermaid diagram indicating how a request from a user triggers a webhook event to an app, who can then review the request and choose to approve it. Later, the application can review all approved PATs, and choose to revoke one, resulting in an email notification to the user who created the PAT.

Details included in the webhook event and API listings include the repositories and permissions requested, the expiration time of the token, and the user's explanation for what they plan to do with the PAT. The personal_access_token_request events are generated when a request is created, approved or denied by an administrator or application, or cancelled by the requesting user.

Only a GitHub app is able to call these APIs, either acting on its own or on behalf of a signed-in organization administrator.
The organization_personal_access_tokens permission is needed to manage the active tokens, while the organization_personal_access_token_requests permission enables the app to recieve webhooks about requests and call the request management APIs.

Organizations must have the personal access token approval flow enabled in order to manage these requests, otherwise fine-grained personal access tokens are automatically approved for the organization (which generates a personal_access_token_request: approved event).

To learn more about these APIs, see "List requests to access organization resources with fine-grained PATs" and the "personal_access_token_request webhook event". To learn more about fine-grained PATs, and how to enable them for your organization, see "Setting a personal access token policy for your organization".

If you have feedback or bugs to report about fine-grained PATs, please let us know in the dedicated feedback discussion.

See more

Fixed bug that allowed private issues and pull request titles to be shown in search results

GitHub Security was notified about an issue where private issue and pull request titles would be displayed in search results. Our Security team investigated potential instances and determined that this only occurred when the author of the commit was authorized to view the issue or pull request and the commit was titled as a link to the private issue or pull request. Additionally, this only happened while using the new code search (beta). This issue was introduced when the new code search (beta) launched and was fixed on January 17, 2023. As this issue has been addressed, there is no further action that is required by any user.

See more

Code scanning API to enable default setup with CodeQL on a repository

Code scanning have shipped an API for repositories to programmatically enable code scanning default setup with CodeQL.

The API can be used to:

  • Onboard a repository to default setup: gh api -X PATCH /repos/[org-name]/[repo-name]/code-scanning/default-setup -f state=configured
  • Specify which CodeQL query suite to use in the default setup configuration: gh api -X PATCH /repos/[org-name]/[repo-name]/code-scanning/default-setup -f query_suite=extended
  • View the current default setup configuration for a repository: gh api /repos/[org]/[repo-name]/code-scanning/default-setup
  • Offboard a repository from default setup: gh api -X PATCH /repos/[org-name]/[repo-name]/code-scanning/default-setup -f state=not-configured

When you onboard a repository via the API, you will recieve a workflow run ID which can be used to monitor the setup progress. This can be used to see the status and conclusion of the run: gh api repos/[org-name]/[repo-name]/actions/runs/[run-id] --jq '.status, .conclusion'

{
  "state": "configured",
  "languages": ["javascript", "ruby"],
  "query_suite": "default", 
  "updated_at": "2023-02-24T20:00:42Z"
}

For more information, see "Get the code scanning default setup configuration" and "Update the code scanning default setup configuration".

See more

General availability of granular access token on npm

Today we are making the granular access token feature on npm generally available.

Granular access token, allows you to:

  • Restrict token access to specific packages and/or scopes
  • Grant tokens access to specific organizations for org and user management
  • Set a token expiration date
  • Limit token access based on IP address ranges
  • Select between read and/or write access for the token

We recommend using granular access tokens with least privileges for automating your publishing and org management activities. You can allow your package to be published without 2FA using granular access tokens from your trusted CI/CD systems. Additionally, you can also configure your package to require 2FA when publishing from a local machine to defend against account hijacking.

Read more about creating a granular access token here.

See more

You can now use the “security extended” query suite in code scanning default setup with CodeQL

You can now enable the "security extended" query suite for repositories using code scanning default setup with CodeQL. This query suite can be selected during set up, or changed at any time by viewing and editing the CodeQL configuration.

Code scanning's default query suites have been carefully designed to ensure that they look for the security issues most relevant to developers, whilst also minimizing the occurrence of false positive results. However, if you and you developers are interested in seeing a wider range of alerts you can enable the security extended query suite. This suite includes the same queries as in the default query suite, plus:

  • extra queries with slightly lower severity and precision.
  • extra experimental queries.

If you enable the security extended suite you may see more CodeQL alerts in your repository and on pull requests. For more information, see "About code scanning alerts".

Code scanning default setup query suites

Code scanning default setup view configuration

Read more about code scanning default setup.

See more

SSH Certificate requirement update

The "Require SSH certificates" policy now allows GitHub apps to call Git APIs using a user-to-server token, bringing them up to parity with OAuth app support.

The SSH certificate requirement mandates that users in your organization call Git APIs using an SSH certificate issued by your organization, in place of their own SSH key or a PAT.
To support automation, it has an exception in place for OAuth apps and GitHub app server-to-server tokens, which allows applications you've approved to call Git APIs for your organization.
With this change, we are extending that exception to GitHub app user-to-server tokens, for when a user has signed into a GitHub app that's installed in your organization.

Screenshot of the SSH Certificate requirement checkbox

This change also applies when the enterprise-level setting requires SSH certificates across all organizations in the enterprise.

To learn more, see "Managing your organization's SSH certificate authorities" or "Managing SSH certificate authorities for your enterprise".

See more

Fixed bug that allowed removed users to retain access to the organization

GitHub Security was notified about an issue where users still had access to organizations after being removed. Our Security team investigated potential instances and determined there were occasional instances where users’ permissions were not fully removed when teams were deleted or they were part of a team when they were removed from the organization. While we investigated the root causes, which stemmed from background job and permissions issues, a manual fix has been implemented since October 20, 2022. We have addressed the underlying issues and the need for the previously implemented manual fix. We have also cleaned up any users that were not removed when they should have been. There is no further action that is required by any organization.

See more

Delete stale code scanning configurations to close outdated alerts

Code scanning configurations can now be deleted from the code scanning alert page. This could be used to delete stale configurations causing alerts to remain open, or delete old configurations which are no longer used.

Code scanning can be configured to use different tools, target different languages, or even analyze different parts of the codebase in the same repository. In certain circumstances more than one of these configurations may produce the same alert. However, if one of the configurations is no longer used and becomes 'stale' you may find that the alert is fixed in one configuration but not in the stale configuration, which is potentially confusing. Today we are releasing a new feature that allows you to easily delete stale configurations which cause alerts to remain open after they've been fixed.

In the code scanning alert page, the counter in the 'Affected branches' sidebar shows the number of configurations for the branch. Click a branch to view the configuration details, and delete configurations as required. A configuration is deleted for a branch, so may have an impact on the status of other alerts on the same branch. When a configuration is deleted, a timeline entry is recorded on the alert, and repositories in an organization also record an audit log entry. If a configuration is deleted by mistake, re-run the analysis to update the alert and reinstate the configuration.

Delete code scanning configurations

Read more about removing stale code scanning configurations and alerts.

See more
View more changes