HCL AION CVE-2025-52635

The specific precondition is that AION 2.0's Content Security Policy does not enforce the Trusted Types mechanism for scripts, removing a browser-level mitigation that would otherwise block unsafe assignments to DOM injection sinks. On its own this is a defense-in-depth gap: realistic exploitation additionally requires an existing injectable DOM sink in the AION 2.0 web interface AND a victim user loading attacker-influenced content in that interface. Limiting factors that reduce real-world risk below the 9.8 label: exploitation is client-side and effectively requires user interaction (loading malicious content in the app), impact is scoped to the browser context of the affected user, and the near-floor EPSS (0.03%) indicates no observed exploitation. Authentication requirements are not independently confirmable - the vendor vector states PR:N, but typical AION usage implies an authenticated application user.

The input presents a sharp conflict between the raw score and the vulnerability class. HCL PSIRT (psirt@hcl.com) assigned CVSS 9.8 with vector AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H (full Confidentiality/Integrity/Availability impact, no privileges, no user interaction), yet the weakness itself - 'Trusted Types in scripts not enforced in CSP' (CWE-1032, a CSP/configuration hardening category) - is a client-side defense-in-depth gap that weakens browser protection against DOM-based XSS rather than a directly remote-code-execution-class flaw. These signals do not cohere: a missing Trusted Types enforcement cannot by itself yield unauthenticated full-triad impact without a separate injectable sink and a victim loading attacker-influenced content, which implies user interaction and higher complexity than AC:L/UI:N suggests. EPSS strongly corroborates the lower real-world risk read: 0.03% probability at the 9th percentile is near the floor, indicating essentially no observed exploitation activity. There is no CISA KEV listing and no public POC referenced. The tag 'Information Disclosure' also points to a more limited consequence than the A:H/I:H in the vendor vector. Net judgment: this is a high-CVSS, low-real-risk item for most environments - a legitimate hardening fix to apply via the vendor patch, but not an emergency relative to its 9.8 label. The vendor CVSS vector appears inflated for a Trusted Types/CSP weakness and should be verified against HCL's advisory.

Because AION 2.0 does not enforce Trusted Types via its Content Security Policy, a DOM-based cross-site scripting sink that would otherwise be blocked remains exploitable: an attacker crafts input or a link that, when rendered in the application for an authenticated user, injects script into a trusted DOM sink. No public exploit code was identified at time of analysis, and realistic exploitation depends on an application user loading the malicious content.