GHSA-XVF4-CH4Q-2M24

Vulnerability from github – Published: 2026-03-16 16:37 – Updated: 2026-03-18 21:51
VLAI?
Summary
StudioCMS REST getUsers Exposes Owner Account Records to Admin Tokens
Details

Summary

The REST API getUsers endpoint in StudioCMS uses the attacker-controlled rank query parameter to decide whether owner accounts should be filtered from the result set. As a result, an admin token can request rank=owner and receive owner account records, including IDs, usernames, display names, and email addresses, even though the adjacent getUser endpoint correctly blocks admins from viewing owner users. This is an authorization inconsistency inside the same user-management surface.

Details

Vulnerable Code Path

File: D:/bugcrowd/studiocms/repo/packages/studiocms/frontend/pages/studiocms_api/_handlers/rest-api/v1/secure.ts, lines 1605-1647

.handle(
    'getUsers',
    Effect.fn(
        function* ({ urlParams: { name, rank, username } }) {
            if (!restAPIEnabled) {
                return yield* new RestAPIError({ error: 'Endpoint not found' });
            }
            const [sdk, user] = yield* Effect.all([SDKCore, CurrentRestAPIUser]);

            if (user.rank !== 'owner' && user.rank !== 'admin') {
                return yield* new RestAPIError({ error: 'Unauthorized' });
            }

            const allUsers = yield* sdk.GET.users.all();
            let data = allUsers.map(...);

            if (rank !== 'owner') {
                data = data.filter((user) => user.rank !== 'owner');
            }

            if (rank) {
                data = data.filter((user) => user.rank === rank);
            }

            return data;
        },

The rank variable in if (rank !== 'owner') is the request query parameter, not the caller's privilege level. An admin can therefore pass rank=owner, skip the owner-filtering branch, and then have the second if (rank) branch return only owner accounts.

Adjacent Endpoint Shows Intended Security Boundary

File: D:/bugcrowd/studiocms/repo/packages/studiocms/frontend/pages/studiocms_api/_handlers/rest-api/v1/secure.ts, lines 1650-1710

const existingUserRankIndex = availablePermissionRanks.indexOf(existingUserRank);
const loggedInUserRankIndex = availablePermissionRanks.indexOf(user.rank);

if (loggedInUserRankIndex <= existingUserRankIndex) {
    return yield* new RestAPIError({
        error: 'Unauthorized to view user with higher rank',
    });
}

getUser correctly blocks an admin from viewing an owner record. getUsers bypasses that boundary for bulk enumeration.

Sensitive Fields Returned

The getUsers response includes:

  • id
  • email
  • name
  • username
  • rank
  • timestamps and profile URL/avatar fields when present

This is enough to enumerate all owner accounts and target them for phishing, social engineering, or follow-on attacks against out-of-band workflows.

PoC

HTTP PoC

Use any admin-level REST API token:

curl -X GET 'http://localhost:4321/studiocms_api/rest/v1/secure/users?rank=owner' \
  -H 'Authorization: Bearer <admin-api-token>'

Expected behavior: - owner records should be excluded for admin callers, consistent with getUser

Actual behavior: - the response contains owner user objects, including email addresses and user IDs

Local Validation of the Exact Handler Logic

I validated the filtering logic locally with the same conditions used by getUsers and getUser.

Observed output:

{
  "admin_getUsers_rank_owner": [
    {
      "email": "owner@example.test",
      "id": "owner-1",
      "name": "Site Owner",
      "rank": "owner",
      "username": "owner1"
    }
  ],
  "admin_getUser_owner": "Unauthorized to view user with higher rank"
}

This demonstrates the authorization mismatch clearly: - bulk listing with rank=owner exposes owner records - direct access to a single owner record is denied

Impact

  • Owner Account Enumeration: Admin tokens can recover owner user IDs, usernames, display names, and email addresses.
  • Authorization Boundary Bypass: The REST collection endpoint bypasses the stricter per-record rank check already implemented by getUser.
  • Chaining Value: Exposed owner contact data can support phishing, account-targeting, and admin-to-owner pivot attempts in deployments that treat owner identities as higher-trust principals.

Recommended Fix

Apply rank filtering based on the caller's role, not on the request query parameter, and reuse the same privilege rule as getUser.

Example fix:

const loggedInUserRankIndex = availablePermissionRanks.indexOf(user.rank);

data = data.filter((candidate) => {
    const candidateRankIndex = availablePermissionRanks.indexOf(candidate.rank);
    return loggedInUserRankIndex > candidateRankIndex;
});

if (rank) {
    data = data.filter((candidate) => candidate.rank === rank);
}

At minimum, replace:

if (rank !== 'owner') {
    data = data.filter((user) => user.rank !== 'owner');
}

with a check tied to user.rank rather than the query parameter.

Show details on source website

{
  "affected": [
    {
      "database_specific": {
        "last_known_affected_version_range": "\u003c= 0.4.3"
      },
      "package": {
        "ecosystem": "npm",
        "name": "studiocms"
      },
      "ranges": [
        {
          "events": [
            {
              "introduced": "0"
            },
            {
              "fixed": "0.4.4"
            }
          ],
          "type": "ECOSYSTEM"
        }
      ]
    }
  ],
  "aliases": [
    "CVE-2026-32638"
  ],
  "database_specific": {
    "cwe_ids": [
      "CWE-639"
    ],
    "github_reviewed": true,
    "github_reviewed_at": "2026-03-16T16:37:42Z",
    "nvd_published_at": "2026-03-18T21:16:26Z",
    "severity": "LOW"
  },
  "details": "## Summary\n\nThe REST API `getUsers` endpoint in StudioCMS uses the attacker-controlled `rank` query parameter to decide whether owner accounts should be filtered from the result set. As a result, an admin token can request `rank=owner` and receive owner account records, including IDs, usernames, display names, and email addresses, even though the adjacent `getUser` endpoint correctly blocks admins from viewing owner users. This is an authorization inconsistency inside the same user-management surface.\n\n## Details\n\n### Vulnerable Code Path\n\nFile: `D:/bugcrowd/studiocms/repo/packages/studiocms/frontend/pages/studiocms_api/_handlers/rest-api/v1/secure.ts`, lines 1605-1647\n\n```ts\n.handle(\n    \u0027getUsers\u0027,\n    Effect.fn(\n        function* ({ urlParams: { name, rank, username } }) {\n            if (!restAPIEnabled) {\n                return yield* new RestAPIError({ error: \u0027Endpoint not found\u0027 });\n            }\n            const [sdk, user] = yield* Effect.all([SDKCore, CurrentRestAPIUser]);\n\n            if (user.rank !== \u0027owner\u0027 \u0026\u0026 user.rank !== \u0027admin\u0027) {\n                return yield* new RestAPIError({ error: \u0027Unauthorized\u0027 });\n            }\n\n            const allUsers = yield* sdk.GET.users.all();\n            let data = allUsers.map(...);\n\n            if (rank !== \u0027owner\u0027) {\n                data = data.filter((user) =\u003e user.rank !== \u0027owner\u0027);\n            }\n\n            if (rank) {\n                data = data.filter((user) =\u003e user.rank === rank);\n            }\n\n            return data;\n        },\n```\n\nThe `rank` variable in `if (rank !== \u0027owner\u0027)` is the request query parameter, not the caller\u0027s privilege level. An admin can therefore pass `rank=owner`, skip the owner-filtering branch, and then have the second `if (rank)` branch return only owner accounts.\n\n### Adjacent Endpoint Shows Intended Security Boundary\n\nFile: `D:/bugcrowd/studiocms/repo/packages/studiocms/frontend/pages/studiocms_api/_handlers/rest-api/v1/secure.ts`, lines 1650-1710\n\n```ts\nconst existingUserRankIndex = availablePermissionRanks.indexOf(existingUserRank);\nconst loggedInUserRankIndex = availablePermissionRanks.indexOf(user.rank);\n\nif (loggedInUserRankIndex \u003c= existingUserRankIndex) {\n    return yield* new RestAPIError({\n        error: \u0027Unauthorized to view user with higher rank\u0027,\n    });\n}\n```\n\n`getUser` correctly blocks an admin from viewing an owner record. `getUsers` bypasses that boundary for bulk enumeration.\n\n### Sensitive Fields Returned\n\nThe `getUsers` response includes:\n\n- `id`\n- `email`\n- `name`\n- `username`\n- `rank`\n- timestamps and profile URL/avatar fields when present\n\nThis is enough to enumerate all owner accounts and target them for phishing, social engineering, or follow-on attacks against out-of-band workflows.\n\n## PoC\n\n### HTTP PoC\n\nUse any admin-level REST API token:\n\n```bash\ncurl -X GET \u0027http://localhost:4321/studiocms_api/rest/v1/secure/users?rank=owner\u0027 \\\n  -H \u0027Authorization: Bearer \u003cadmin-api-token\u003e\u0027\n```\n\nExpected behavior:\n- owner records should be excluded for admin callers, consistent with `getUser`\n\nActual behavior:\n- the response contains owner user objects, including email addresses and user IDs\n\n### Local Validation of the Exact Handler Logic\n\nI validated the filtering logic locally with the same conditions used by `getUsers` and `getUser`.\n\nObserved output:\n\n```json\n{\n  \"admin_getUsers_rank_owner\": [\n    {\n      \"email\": \"owner@example.test\",\n      \"id\": \"owner-1\",\n      \"name\": \"Site Owner\",\n      \"rank\": \"owner\",\n      \"username\": \"owner1\"\n    }\n  ],\n  \"admin_getUser_owner\": \"Unauthorized to view user with higher rank\"\n}\n```\n\nThis demonstrates the authorization mismatch clearly:\n- bulk listing with `rank=owner` exposes owner records\n- direct access to a single owner record is denied\n\n## Impact\n\n- **Owner Account Enumeration:** Admin tokens can recover owner user IDs, usernames, display names, and email addresses.\n- **Authorization Boundary Bypass:** The REST collection endpoint bypasses the stricter per-record rank check already implemented by `getUser`.\n- **Chaining Value:** Exposed owner contact data can support phishing, account-targeting, and admin-to-owner pivot attempts in deployments that treat owner identities as higher-trust principals.\n\n## Recommended Fix\n\nApply rank filtering based on the caller\u0027s role, not on the request query parameter, and reuse the same privilege rule as `getUser`.\n\nExample fix:\n\n```ts\nconst loggedInUserRankIndex = availablePermissionRanks.indexOf(user.rank);\n\ndata = data.filter((candidate) =\u003e {\n    const candidateRankIndex = availablePermissionRanks.indexOf(candidate.rank);\n    return loggedInUserRankIndex \u003e candidateRankIndex;\n});\n\nif (rank) {\n    data = data.filter((candidate) =\u003e candidate.rank === rank);\n}\n```\n\nAt minimum, replace:\n\n```ts\nif (rank !== \u0027owner\u0027) {\n    data = data.filter((user) =\u003e user.rank !== \u0027owner\u0027);\n}\n```\n\nwith a check tied to `user.rank` rather than the query parameter.",
  "id": "GHSA-xvf4-ch4q-2m24",
  "modified": "2026-03-18T21:51:20Z",
  "published": "2026-03-16T16:37:42Z",
  "references": [
    {
      "type": "WEB",
      "url": "https://github.com/withstudiocms/studiocms/security/advisories/GHSA-xvf4-ch4q-2m24"
    },
    {
      "type": "ADVISORY",
      "url": "https://nvd.nist.gov/vuln/detail/CVE-2026-32638"
    },
    {
      "type": "WEB",
      "url": "https://github.com/withstudiocms/studiocms/commit/aebe8bcb3618bb07c6753e3f5c982c1fe6adea64"
    },
    {
      "type": "PACKAGE",
      "url": "https://github.com/withstudiocms/studiocms"
    },
    {
      "type": "WEB",
      "url": "https://github.com/withstudiocms/studiocms/releases/tag/studiocms@0.4.4"
    }
  ],
  "schema_version": "1.4.0",
  "severity": [
    {
      "score": "CVSS:3.1/AV:N/AC:L/PR:H/UI:N/S:U/C:L/I:N/A:N",
      "type": "CVSS_V3"
    }
  ],
  "summary": "StudioCMS REST getUsers Exposes Owner Account Records to Admin Tokens"
}


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…