EnergizeLab Station Middleware CVE-2026-105039

Exploitation requires (1) an affected deployment running EnergizeLab station-middleware Docker image versions 1.6.5 through 1.8.2, and (2) a valid authenticated account of ANY role - the flaw is that administrative API endpoints perform no role/authorization check, so even the lowest-privilege authenticated user qualifies (consistent with PR:L). What LIMITS exploitation: the attacker must already possess authenticated credentials or a session (not pre-auth), and must have network reachability to the middleware's API; versions outside 1.6.5-1.8.2 are not in scope. No special non-default feature toggle is described - the missing authorization applies to the administrative APIs as shipped.

The input CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:N/S:C/C:H/I:H/A:N, 9.6 Critical) is internally consistent with the description: a CWE-862 missing-authorization flaw where administrative NestJS APIs omit role checks, so any authenticated low-privilege user crosses a security boundary (S:C) to read secrets (C:H) and tamper with config/object-store (I:H), with no availability impact (A:N). The PR:L requirement is the single most important risk-reducing signal - this is NOT remote unauthenticated exploitation; an attacker first needs a valid low-privilege account (or stolen/phished credentials, or an existing malicious tenant). No EPSS score was provided, no CISA KEV listing is indicated, and no public exploit can be confirmed from the references (see confidence_notes), so the three independent real-world-exploitation signals are all absent or unverifiable. Net: for organizations actually running EnergizeLab station-middleware this is a genuine high-priority patch because the barrier (any authenticated role) is trivially low and the loot (AI provider keys) is high-value; but at internet scale the risk is tempered by an obscure product footprint and the lack of EPSS/KEV/POC corroboration, pointing to targeted rather than widespread exploitation. Missing data to verify: EPSS probability, KEV status, and whether the GitHub reference is an actual working exploit.

An attacker holding any valid low-privilege account on an affected EnergizeLab station-middleware instance (versions 1.6.5-1.8.2) directly invokes the administrative API endpoints, which never validate the caller's role. The attacker dumps the stored AI provider API keys and rewrites configuration and object-store contents. No public, confirmed exploit was identified at time of analysis, though one reference is a GitHub repository of unverified content.