Sling Security Bundle CVE-2026-94251

Exploitation requires a deployment that actually relies on Apache Sling's ContentDispositionFilter (Security Bundle before 1.3.12) to force attachment/download handling on attacker-influenceable resources; the attacker must be able to place or influence such a resource, then access it through a second URL/API shape of the same resource that the filter does not mediate, and a victim must open that crafted link (user interaction). It does not affect deployments that do not depend on this filter for content-disposition protection or that have upgraded to 1.3.12. No CVSS vector is provided, so authentication level cannot be confirmed from the data.

This is a protection-mechanism-failure (CWE-693) in Apache Sling's ContentDispositionFilter, which normally forces a 'Content-Disposition: attachment' header to prevent user-uploaded/binary resources from being rendered inline in the browser (a stored-XSS/drive-by defense). The bug: the filter mediates only one address/API shape of a resource, so the same resource fetched via an alternate URL or API shape escapes the filter and is served inline. No CVSS vector, EPSS score, KEV listing, or public POC is provided, so severity signals are entirely qualitative. This is a defense-in-depth bypass rather than a direct memory-corruption or auth flaw: real impact depends on whether the deploying application actually relies on ContentDispositionFilter to neutralize attacker-influenced content, making it a moderate, conditional issue rather than a drop-everything priority. Because no CVSS is supplied, authentication requirements cannot be confirmed from the vector; the alternate-shape access is likely low-complexity but the payoff (inline rendering → XSS/content-sniffing) requires a victim to open the crafted URL. Treat as a patch-in-normal-cycle item, prioritized higher for Adobe AEM / Sling deployments that host untrusted user content.

An attacker stores or influences content on a Sling/AEM site that would normally be forced to download as an attachment, then crafts a request that reaches the same resource through a different URL or API shape not covered by ContentDispositionFilter. When a victim opens that link, the browser renders the content inline (e.g., an HTML/SVG payload), enabling stored/reflected XSS or MIME-sniffing in the victim's session. No public exploit code is identified at time of analysis.