GHSA-c4cm-r9fh-jgj9 json
commonground-api-common unexploitable privilege escalation in JWT authentication middleware
published 2y ago modified last yr.
- cvss
- —
- not scored
- cisa kev
- Not listed
- not in CISA's catalog
- epss
- —
- no score, EPSS covers CVEs only
- public exploits
- 0
- none known
Summary
### Impact
This is a privilege escalation vulnerability. The impact is negligible and entirely theoretical.
A non-exploitable weakness was found in how the client-supplied JWTs are verified. Because an explicit allow-list of known algorithms is used in the PyJWT library, user-supplied (invalid) algorithms are rejected.
If this was not the case, then the client JWTs could be tampered with, resulting in privilege escalation which would allow the attacker to perform any operation as any client (impersonation) without leaving a trace of the real user/client.
### Patches
Will be fixed in 1.12.2
### Workarounds
None needed. But be careful when updating PyJWT. Check that the used PyJWT has no algorithms specified with a name in "", "HS25", "HS2", "HS", "H", or that those algorithms are acceptable.
### Details
The header and payload of JSON Web Tokens (JWTs) are cryptographically signed with an algorithm. A JWT has a header field `alg` that specifies the algorithm used in the signature.
The `vng-api-common.middleware.AuthMiddleware` uses PyJWT to check the validity of JWT and indicates it should be "HS256", otherwise an attacker could construct a token with a cryptographically weak token. It should indicate this with a list of acceptable algorithms `["HS256"]`, but instead the string `"HS256"` is passed to PyJWT. PyJWT does not check the type of the argument and checks if the `alg` string in the header exists in the acceptable algorithms value with the `in` operator. Any substring of `"HS256"` passes this `in` check. It is not exploitable because there is no such substring in de set of algorithms PyJWT supports.
Patch / fix
Is GHSA-c4cm-r9fh-jgj9 in your dependencies? Fencer shows which of your repositories are affected as soon as an advisory lands, and opens the fix as a pull request.
Check with FencerAffected packages
| ecosystem | package | affected | fixed in |
|---|---|---|---|
| pypi | commonground-api-common | <=1.12.1 |
no fix listed |
| pypi | vng-api-common | <=2.0.5 |
no fix listed |
| pypi | vng-api-common-utrecht | <=1.3.2 |
no fix listed |
References
Sources
- GitHub Advisory GHSA-c4cm-r9fh-jgj9
Fencer Intel keeps one entry per vulnerability: the CVE id wins, then GHSA, and records that alias each other are merged, except that two CVEs always keep separate entries. The same data is available as JSON.