The main possibility for security bugs in cagebreak is privilege escalation via the socket. Any program with access to the socket immediately gains arbitrary code execution rights. The socket has to be explicitely enabled using the -e flag on invocation and is restricted to the user of the cagebreak process (700).

If you want to get in touch with project-repo via email, contact cagebreak @ project-repo . co.

We try to respond to everything that is not obvious spam.

If you can, please encrypt your email with the appropriate GPG key found in keys/ and sign your message with your own key.

• B15B92642760E11FE002DE168708D42451A94AB5 (expired)
• F8DD9F8DD12B85A28F5827C4678E34D2E753AA3C (expired)
• 3ACEA46CCECD59E4C8222F791CBEB493681E8693 (expired)
• 0A268C188D7949FEB39FD1462F2AD980247E4918 (soon to expire)
• 283D10F54201B0C6CCEE2C561DE04E4B056C749D

Note that our keys are signed by cagebreak signing keys.

If you want us to respond via GPG-encrypted email, please include your own public key or provide the fingerprint and directions to obtain the key.

The most recent release always contains the latest bug fixes and features. There are no official backports for security vulnerabilities. Builds are reproducible under conditions outlined in README.md.

For normal bugs you may open an issue on github.

For everything else, an email contact (with gpg encryption and signature) is available above.

Cagebreak is a wayland compositor run by a user and has access to the resources the user has access to. Cagebreak cannot restrict other programs (consider a web browser unable to write a downloaded file for instance).

There is no transmission of information by cagebreak other than to the screens, ipc (if enabled with -e) and potentially other documented local channels.

This is not a thorough analysis, just an overview of the ways in which cagebreak has (no) attack surface. Please reference the man pages (especially options -e and --bs ).

Not applicable - Cagebreak is executed after user login.

Not applicable - Cagebreak allows system manipulation for user software.

Not applicable - There are no prohibited operations (See Tampering above.). Cagebreak sends events over documented channels. There is no logging activated by default - this can be changed by logging the socket for example.

Not applicable - Information disclosure over documented channels is a feature. Any user software may exfiltrate any data the user has access to.

Not applicable - Cagebreak offers functionality to terminate itself. This is available to user software over the socket.

Software may gain arbitrary code execution if it has access to the socket. Privilege escalation to root is unlikely since privileges are dropped before any user input is accepted.

All Cagebreak project keys are found under keys/ in the cagebreak repository (the public keys anyway).

The most trusted keys of the Cagebreak project are its signing keys, all signing keys are signed by at least one of its predecessors and at least one non-expired signing key is used at the time of release to sign the commit tag and the release code tarball.

Signing keys are also used to lend credence to other keys in the Cagebreak project, such as the keys for email correspondence and the key used in the cagebreak-pkgbuild repository.