CVE-2026-12143

A flaw was found in form-data, a library for creating readable multipart/form-data streams. A remote attacker can exploit this vulnerability by injecting carriage return (CR), line feed (LF), or double-quote (") characters into the field argument of FormData#append or the filename option. This allows the attacker to inject additional headers or smuggle entire additional multipart parts into requests, potentially enabling them to add or override form fields and compromise data integrity.

This is an Important impact flaw in the form-data library: a remote attacker can inject arbitrary headers or additional multipart parts via CRLF injection in field names or filenames, potentially overriding sensitive form fields and affecting data integrity.

For RHOAI and RHEL AI, severity is Moderate because affected versions appear only as a transitive npm dependency in RHOAI (dashboard, mod-arch plugins, MLflow UI) and RHEL AI 3.4 bootc images, and those products use fixed field names for uploads rather than passing untrusted user input as multipart field names or filenames. The documented exploit path is therefore not reachable in default deployments. Practical impact is limited to non-default or custom integrations that forward multipart requests using attacker-controlled field names.

The CVSS score(s) listed for this vulnerability do not reflect the associated product's status, and are included for informational purposes to better understand the severity of this vulnerability.

Applications using the form-data library should implement strict input validation and sanitization for all field names and filenames derived from untrusted sources. This prevents the injection of control characters (CR, LF, ") that could lead to header injection or form field overrides. Deployments that exclusively use fixed or trusted field names are not impacted.