harden-runner VS repo

Fortunately there is a great free online tool that help you by doing all the hard work (it will open a pull-request and automatically fix issues).

If using GitHub Actions for CI/ CD, Harden Runner (https://github.com/step-security/harden-runner) can be used to audit and block DNS exfiltration. Outbound calls from CI are predictable (to source repo, artifact registry, etc.) and don't change often.

As part of writing tests for Harden Runner GitHub Action, which prevents such attacks, there was a need to write attack simulator for these attacks.

I am not too familiar with GitLab, to be honest, but: - Commit/PR linting (to be in tandem with semantic versioning) is implemented via third-party GitHub Actions (https://github.com/amannn/action-semantic-pull-request and https://github.com/wagoid/commitlint-github-action), these might be hard to transfer - Blocking egress to mitigate supply chain attacks is performed by step security’s Harden Runner (https://github.com/step-security/harden-runner), you may raise a question there about GitLab support - CodeQL support is GitHub only AFAIK (but you would have to verify it)

I've found StepSecurity's tooling helpful in getting my repos secured.

* https://app.stepsecurity.io/securerepo

I agree. There are projects such as https://github.com/ossf/package-analysis and https://github.com/step-security/harden-runner that do behavior analysis. Disclaimer: I’m maintainer of the second one.

So if you are concerned about this, I'd suggest looking at the following:

* OpenSSF Scorecard Action - https://github.com/ossf/scorecard#scorecards-github-action

* Step Security Harden Action - https://github.com/step-security/harden-runner

I realize that this means trusting these providers but they seem at least tacitly blessed by GitHub. https://docs.github.com/en/actions/security-guides/security-...

Few hours back several malicious packages were released on npm registry. This video shows how some of these packages are making outbound calls as part of the preinstall step when executed in a GitHub Actions workflow. DNS Exfiltration and network calls detected by Harden-Runner GitHub Action https://github.com/step-security/harden-runner

# download a repository on GitHub to our machine # Replace `owner/repo` with the owner and name of the repository to clone git clone https://github.com/owner/repo.git # change into the `repo` directory cd repo # create a new branch to store any new changes git branch my-branch # switch to that branch (line of development) git checkout my-branch # make changes, for example, edit `file1.md` and `file2.md` using the text editor # stage the changed files git add file1.md file2.md # take a snapshot of the staging area (anything that's been added) git commit -m "my snapshot" # push changes to github git push --set-upstream origin my-branch

// https://github.com/owner/repo/blob/branch/path/to/Contract.sol import "https://github.com/owner/repo/blob/branch/path/to/Contract.sol"; // Example import ERC20.sol from openzeppelin-contract repo // https://github.com/OpenZeppelin/openzeppelin-contracts/blob/master/contracts/token/ERC20/ERC20.sol import "https://github.com/OpenZeppelin/openzeppelin-contracts/blob/master/contracts/token/ERC20/ERC20.sol";

Thanks for the feedback!

w.r.t Bypass idea 1: in the future, the plan is to have application level policy, e.g. only allow 'GET https://github.com/owner/repo'.

w.r.t Bypass idea 2: in the future, the plan is to have a policy to disable sudo. Once that is done, user code will not be able to mess with the agent.

GCM Core does support multiple users with GitHub remotes. You have to put the username in the remote URL, for example:

https://[email protected]/owner/repo

GCM will store credentials for "alice" only. Repositories that have remote URLs with a different "user@" will use different credentials.

Remote URLs without a "user@" part at all (like "https://github.com/owner/repo") are treated like the "default" user for GitHub.com, rather than a specific one.