Code injection in Orval via unsanitized x-enum-descriptions
CVE-2026-25141 is a critical code-injection vulnerability in Orval, a tool that generates type-safe JavaScript/TypeScript clients from OpenAPI v3 and Swagger v2 specifications. The issue affects versions starting with 7.19.0 and prior to 7.21.0, and 8.x prior to 8.2.0, and is described as an incomplete fix bypass for CVE-2026-23947. The vulnerable behavior occurs when Orval embeds attacker-controlled x-enum-descriptions from an OpenAPI document into generated TypeScript/JavaScript output as comments without fully sanitizing dangerous characters. Reported details indicate that the sanitization logic does not adequately neutralize characters such as '' and '/' in comment context, allowing an attacker to inject the sequence '/' to terminate a comment early and append attacker-controlled code. Additional reporting notes that even with quote escaping in jsStringEscape, arbitrary JavaScript injection remains possible using only a restricted character set such as !+ via JSFuck-style payloads. If a developer or CI pipeline generates code from a malicious or compromised OpenAPI specification, the injected content can be written into generated source files and subsequently executed during build, test, or runtime.
What it means and what to do now: patch path, interim mitigations, and the assume-compromise checklist.
What an attacker gets, and what they’ve been doing with it.
Successful exploitation allows arbitrary JavaScript/TypeScript code injection into generated client code. Depending on how the generated artifacts are used, this can lead to execution of attacker-controlled code in developer workstations, CI/CD runners, or application environments under the privileges of the affected process. Consequences can include theft of environment variables, API keys, tokens, source code, or user data; tampering with generated clients; supply-chain compromise of downstream builds; and broader compromise of systems that execute the generated code.
If you can’t patch tonight, do this now.
Until patching is complete, do not generate clients from untrusted or externally supplied OpenAPI/Swagger specifications. Treat API specifications as untrusted input, restrict who can modify them, pin and review spec sources, and require code review for spec changes. Run code generation in isolated, sandboxed environments with minimal filesystem, network, and secret access. Avoid executing generated code automatically in privileged build stages, and monitor generated files for suspicious comment breakouts or unexpected code blocks.
Patch, then assume compromise.
Upgrade Orval/@orval/core to a fixed release: 7.21.0 or later in the 7.x line, or 8.2.0 or later in the 8.x line. Remove vulnerable versions from developer environments and CI pipelines, regenerate affected clients from trusted specifications after upgrading, and review generated artifacts for unexpected injected code. Audit build and dependency management workflows to ensure no vulnerable Orval versions remain in use.
No public exploits tracked yet. Mallory keeps watching.
Products and vendors Mallory has correlated with this vulnerability. Open in Mallory to drill down to specific CPE configurations and version ranges.
7 sources tracked across advisories, community write-ups, and news. New activity surfaces here as Mallory finds it.