Six Proto6 Vulnerabilities in protobuf.js Expose Node.js Apps to RCE and DoS
Six flaws in protobuf.js can crash Node.js apps and, in some cases, lead to remote code execution. Patches are available for protobuf.js and protobufjs-cli.
Intelligence analysis by GPT-5.4 Mini
Cyera says the Proto6 bugs arise from protobuf.js trusting schemas and metadata too much. The result can be DoS, code injection, or full RCE in Node.js services, CI/CD pipelines, and other systems that process Protobuf data.
A helper tool in Node.js trusted labels on data boxes too much. If a bad person slipped in a fake label, the app could freeze, break, or even run the wrong code, like a machine that follows a trick instruction.
Analysis
What was found
Cybersecurity researchers flagged six vulnerabilities in protobuf.js, the JavaScript and TypeScript implementation of Protocol Buffers. The issues were grouped under the name Proto6 and span denial-of-service bugs, prototype pollution-style behavior, and code injection paths.
The article says affected environments include Node.js applications that deserialize Protobuf data or generate code from schemas, plus Google Cloud client libraries, messaging frameworks such as Baileys, and CI/CD pipelines. Cyera’s assessment is that the library treats schemas and metadata as trusted by default, and that this trust model is what opens the door to exploitation.
Severity and attack paths
The most severe issue described is CVE-2026-44291, rated 8.1. In the article’s explanation, attacker-controlled input can reach a prototype pollution gadget, and later protobuf.js can use polluted property lookups while encoding or decoding a message. That can cause generated code to include attacker-controlled strings and then compile them with Function(), which leads to arbitrary JavaScript execution inside the Node.js process.
Other flaws can trigger denial-of-service conditions through unbounded recursion, unsafe option paths, crafted field names, or prototype injection in generated constructors. One case, CVE-2026-44295, can lead to code injection in pbjs static output when schema names are crafted maliciously.
Fixes and exposure
The article lists vulnerable versions of protobuf.js as <= 7.5.5 and 8.0.0 through 8.0.1, with protobufjs-cli affected at <= 1.2.0 and 2.0.0 through 2.0.1. Fixed releases are protobufjs 7.5.6 and 8.0.2, plus protobufjs-cli 1.2.1 and 2.0.2.
Cyera warns that the risk matters beyond classic app code because many modern pipelines treat schemas, configuration files, and metadata as inputs that drive automation and code generation. In that setup, a malicious schema can turn data handling into behavior-changing execution.
Key points
- Researchers identified six vulnerabilities in protobuf.js, labeled Proto6.
- The bugs can cause denial of service and, in some cases, remote code execution.
- Affected software includes Node.js apps, Google Cloud client libraries, Baileys, and CI/CD pipelines.
- Fixed releases are available for protobuf.js and protobufjs-cli.
- The root problem is that the library trusts schemas and metadata too much by default.
Patches are already available for the affected protobuf.js and protobufjs-cli versions, so teams can reduce exposure quickly by upgrading. The report may also push developers to treat schemas and metadata as untrusted inputs, which could improve security in build systems and data pipelines.
If old versions stay in services or CI/CD tooling, a crafted schema or payload could still cause crashes or code execution. The article also suggests that systems built around shared schemas and automation may be especially risky when trust assumptions are wrong.
