Cloud Foundry UAA CVE-2026-47839

Exploitation requires the target UAA to have a federated OIDC identity provider configured with groupMappingMode: AS_SCOPES AND a wildcard ('*') entry in that provider's externalGroupsWhitelist - this is the explicit prerequisite stated in the description and is not the default configuration. The attacker must be able to authenticate through that federated OIDC provider (i.e. be a valid user of the external IdP, or control/influence the group claims it asserts). Limiting factors: the specific AS_SCOPES + wildcard combination is non-default (reflected as AC:H/AT:P in the CVSS 4.0 vector); deployments using explicit group whitelists or internal group mapping are not exploitable.

The CVSS 4.0 score of 9.2 reflects network attack vector (AV:N), no privileges on UAA itself (PR:N), no user interaction (UI:N), and full compromise of confidentiality, integrity and availability of the UAA control plane (VC:H/VI:H/VA:H) - obtaining uaa.admin effectively grants control of identity and token issuance across the foundation. Crucially the vector also encodes AC:H and AT:P, meaning exploitation is not trivial: it depends on a specific present condition (the AS_SCOPES + wildcard whitelist configuration), which is not UAA's default. There is no EPSS or KEV signal supplied and no POC referenced, so there is no evidence of active exploitation - this is a high-impact-but-conditionally-exploitable issue. For environments that happen to run the vulnerable configuration the real-world risk is severe (full admin takeover); for everyone else it is latent. The input 'Information Disclosure' tag understates the impact, which is privilege escalation to platform admin.

An enterprise federates an external OIDC identity provider into UAA using groupMappingMode: AS_SCOPES and sets externalGroupsWhitelist to '*' for convenience. A user in the external directory (or someone who can influence their group claims) authenticates through that provider and receives a token carrying uaa.admin, gaining administrative control over UAA and the Cloud Foundry foundation. No public exploit code is identified at time of analysis; the attack relies on the misconfiguration being present.