Cloud Pak Cyclops CVE-2025-36220

Exploitation requires an authenticated session with at least low-privilege access, as confirmed by the CVSS vector PR:L - unauthenticated exploitation is explicitly not supported. The attack is network-accessible (AV:N) with low complexity (AC:L), meaning no special network position, race condition, or chained vulnerability is needed beyond valid credentials. The targeted system must be running IBM Cloud Pak for Data System - Cyclops in the version range 11.3.0.2 through Interim Fix 002. SSVC rates this as not automatable, indicating the vulnerable input handling likely requires structured, context-aware requests rather than trivially mass-exploitable network scanning. The primary factor limiting real-world risk is the low-privilege authentication prerequisite, which constrains the attacker pool to users who already have some form of legitimate or compromised system access.

The CVSS 4.3 Medium score reflects a constrained impact profile: network-accessible (AV:N), low complexity (AC:L), low privileges required (PR:L), no user interaction (UI:N), unchanged scope (S:U), no confidentiality impact (C:N), limited integrity impact (I:L), and no availability impact (A:N). However, the vendor description states the attacker may also view data, suggesting the C:N assignment may underrepresent actual impact - this conflict should be verified with IBM. EPSS at 0.03% (9th percentile) indicates the vulnerability is far below the exploitation probability of most actively targeted flaws. SSVC corroborates this: exploitation is rated none, the attack is not automatable, and technical impact is partial. There is no CISA KEV listing and no known public exploit code. Combined, these signals classify this as a routine patch-cycle item rather than an urgent emergency response priority, provided access controls already limit the pool of authenticated low-privilege users.

An attacker who has obtained low-privilege credentials to IBM Cloud Pak for Data System - Cyclops - whether through phishing, credential reuse, or a prior compromise - authenticates to the application and submits a crafted HTTP request containing SQL meta-characters or injected subqueries targeting a vulnerable input field. The back-end database processes the altered query, returning unauthorized records or permitting data modification without additional privilege escalation. No public exploit code has been identified at time of analysis, so exploitation currently requires manual crafting of SQL payloads.