CVE-2026-111371: SchemaEngine AI <= 3.1.3 - Authenticated (Contributor+) Stored Cross-Site Scripting via 'post_content' and 'post_excerpt' Parameters
The SchemaEngine AI – AI Schema Markup, Reviews & Rich Snippets for SEO plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the 'content/excerpt (REST wp/v2/posts) → stored in postcontent/postexcerpt' parameter in all versions up to, and including, 3.1.3 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with contributor-level access and above, to inject arbitrary web scripts in pages that will execute whenever a user accesses an injected page. A Contributor's entity-encoded payload survives save-time wpkses filtering verbatim and is only decoded into raw script tags at render time via htmlentitydecode(), meaning the injected script executes for every visitor — including authenticated Administrators — once an Editor publishes the post through the standard Contributor workflow.
Affected Software
Event History
Frequently Asked Questions
Which sites and users are exposed to exploitation?
WordPress sites running SchemaEngine AI versions through 3.1.3 are affected. Exploitation requires an authenticated account with Contributor-level access or higher, and the payload executes for visitors to an injected page after an Editor publishes it.
Does a Contributor need permission to publish posts directly?
No. The described attack follows the standard Contributor workflow: a Contributor saves the malicious content, and an Editor publishes the post. Once published, the script executes for every visitor, including authenticated Administrators.
What input paths should be reviewed for malicious content?
Review post_content and post_excerpt for posts created or edited through the REST wp/v2/posts content/excerpt fields. Entity-encoded script payloads are particularly relevant because they can survive save-time filtering and are decoded when the page is rendered.
What is the impact if an Administrator views an injected page?
The stored script executes in the Administrator's browser when they access the published injected page. The vulnerability is therefore not limited to the Contributor account that submitted the content.