GHSA-PPPJ-HQ3G-57PJ

Vulnerability from github – Published: 2026-07-22 23:16 – Updated: 2026-08-12 19:58
VLAI
Summary
JupyterLab: Cross-site scripting (XSS) via crafted settings file (`overrides.json`)
Details

JupyterLab 4.5+ allows notebook settings to be shared and applied through an overrides.json file using the Import button in the Settings Editor.

Certain notebook display settings were not properly validated before being applied. As a result, a crafted settings file could contain hidden instructions that run as code inside JupyterLab when imported, instead of only changing a display preference.

Because importing a settings file appears harmless, a user could import a file shared by another party without realizing it could do more. On multi-tenant file systems without proper permission control, another user could plant a malicious overrides.json.

CVE assignment pending, GitHub CNA is experiencing severe backlog

Impact

When a malicious settings file is applied, the embedded code runs with the same access as the affected user. This could allow an attacker to read or modify that user's notebooks and files, and to run code on the user's behalf through the notebook server, including on any connected kernel.

User Interaction vs Privileges Required

Write access to a loaded settings location

If an attacker can write to a directory JupyterLab loads settings from (e.g. on shared or multi-tenant file system), they could place a crafted overrides.json that is applied to another user automatically at startup. This requires high privilages but no action by the victim.

User-imported settings file

A user can import a crafted overrides.json through the Import button in the Settings Editor, having received it from another party. This requires no privileges but a deliberate action by the victim, who reasonably expects a settings file to change preferences rather than run code.

Patches

JupyterLab 4.6.2 and 4.5.10 were patched.

Workarounds

None

Hardening

  1. Treat a settings file as something that can affect how JupyterLab behaves, not only how it appears. Administrators are encouraged to establish a trusted process for distributing configuration rather than relying on ad-hoc importing of shared files.
  2. On multi-tenant or shared file systems, restrict write permissions on the application settings directory and other Jupyter configuration paths so that one user cannot place an overrides.json (or other configuration) readable by another user. A settings file in these locations is applied automatically, without an import step, so directory permissions are the primary control against cross-user tampering.
Severity
8.6 (High)
CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:L/SC:N/SI:N/SA:L
Show details on source website
To clipboard 

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 4.6.1"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "jupyterlab"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "4.6.0"
            },
            {
              "fixed": "4.6.2"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    },
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 4.5.9"
      },
      "package": {
        "ecosystem": "PyPI",
        "name": "jupyterlab"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "3.3.0"
            },
            {
              "fixed": "4.5.10"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-73417"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-79",
      "CWE-116"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-07-22T23:16:18Z",
    "nvd_published_at": null,
    "severity": "HIGH"
  },
  "details": "JupyterLab 4.5+ allows notebook settings to be shared and applied through an `overrides.json` file using the `Import` button in the Settings Editor.\n\nCertain notebook display settings were not properly validated before being applied. As a result, a crafted settings file could contain hidden instructions that run as code inside JupyterLab when imported, instead of only changing a display preference.\n\nBecause importing a settings file appears harmless, a user could import a file shared by another party without realizing it could do more. On multi-tenant file systems without proper permission control, another user could plant a malicious `overrides.json`.\n\n\u003e CVE assignment pending, GitHub CNA is experiencing severe backlog\n\n### Impact\n\nWhen a malicious settings file is applied, the embedded code runs with the same access as the affected user. This could allow an attacker to read or modify that user\u0027s notebooks and files, and to run code on the user\u0027s behalf through the notebook server, including on any connected kernel.\n\n#### User Interaction vs Privileges Required\n\n##### Write access to a loaded settings location\n\nIf an attacker can write to a directory JupyterLab loads settings from (e.g. on shared or multi-tenant file system), they could place a crafted `overrides.json` that is applied to another user automatically at startup. This requires high privilages but no action by the victim. \n\n##### User-imported settings file\n\nA user can import a crafted `overrides.json` through the `Import` button in the Settings Editor, having received it from another party. This requires no privileges but a deliberate action by the victim, who reasonably expects a settings file to change preferences rather than run code.\n\n### Patches\n\nJupyterLab 4.6.2 and 4.5.10 were patched.\n\n### Workarounds\n\nNone\n\n### Hardening\n\n1. Treat a settings file as something that can affect how JupyterLab behaves, not only how it appears. Administrators are encouraged to establish a trusted process for distributing configuration rather than relying on ad-hoc importing of shared files.\n2. On multi-tenant or shared file systems, restrict write permissions on the application settings directory and other Jupyter configuration paths so that one user cannot place an `overrides.json` (or other configuration) readable by another user. A settings file in these locations is applied automatically, without an import step, so directory permissions are the primary control against cross-user tampering.",
  "id": "GHSA-pppj-hq3g-57pj",
  "modified": "2026-08-12T19:58:18Z",
  "published": "2026-07-22T23:16:18Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/security/advisories/GHSA-pppj-hq3g-57pj"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/pull/19184"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/pull/19185"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/pull/19186"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/commit/be9303f5bcd5308eaeae953c5a3c903046682c2c"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/commit/f1beab4a2027af4719d6edc07d52d6cf5a39a432"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/jupyterlab/jupyterlab"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/releases/tag/v4.5.10"
    },
    {
      "type": "WEB",
      "url": "https://github.com/jupyterlab/jupyterlab/releases/tag/v4.6.2"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:4.0/AV:N/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:L/SC:N/SI:N/SA:L",
      "type": "CVSS_V4"
    }
  ],
  "summary": "JupyterLab: Cross-site scripting (XSS) via crafted settings file (`overrides.json`)"
}



Log in or create an account to share your comment.




Tags
Taxonomy of the tags.





Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.

Sightings

Author Source Type Date Other

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.


Detection rules are retrieved from Rulezet.