Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessAuthenticate as low-priv query user
-
ExploitCraft polymorphic join where-filter on hidden field
-
ExecutionInfer restricted values via comparison operators
-
ImpactExtract password-reset token for account takeover
Vulnerability Assessment
| Weakness |
CWE
OWASP Top 10:2025
A01:2025 - Broken Access Control
ASVS 5.0
V14 - Data Protection Data exposure prevention
MITRE ATT&CK
2 techniques across 2 tactics: Collection, Discovery The technique list with ATT&CK links and the kill-chain view is part of the Basic plan and up. |
| Exploitation Conditions | Requires an authenticated user holding query access (PR:L) to a Payload collection that is configured with a polymorphic join relationship referencing a collection containing hidden or read-restricted fields (e.g. password-reset tokens); the attacker abuses the polymorphic join 'where' filter, which prior to the fix failed to validate unknown field paths. Exploitation is limited by: the need for a valid authenticated account, the requirement that a polymorphic join to sensitive fields actually exists in the schema, and the affected version window (Payload >=3.0.0 <3.90.0 or >=4.0.0-canary.0 <4.0.0-canary.34). There is no complete workaround; restricting read access to sensitive collections only reduces exposure, so patching is required. |
| Impact | This is a genuine but moderate-priority information-disclosure flaw (CWE-200) in Payload CMS. The vendor-supplied CVSS 4.0 vector (AV:N/AC:L/AT:N/PR:L/UI:N/VC:H) is internally consistent with the description: it is remotely reachable over the normal query API, low attack complexity, requires authenticated low-privilege access (PR:L), needs no user interaction, and has high confidentiality impact with no integrity or availability impact - matching the leak of hidden/read-restricted values including password-reset tokens. No EPSS score, KEV listing, or public POC was provided in the intelligence set, so there is no evidence of active exploitation or weaponization; this is a high-confidentiality-impact bug without a confirmed exploitation signal. The practical risk is tempered by two prerequisites that are NOT reflected in the raw score: the attacker must already hold query access (an authenticated/low-priv account), and the target collection must expose a polymorphic join to sensitive fields. The password-reset-token angle is the most concerning real-world impact, since inferring a reset token can enable downstream account takeover even though this CVE itself only scores confidentiality. Treat as a should-patch-soon issue for exposed/multi-tenant Payload deployments rather than an emergency, and note the signal gap: no exploitation probability data was supplied, so prioritization leans on impact and ease-of-access rather than observed exploitation. |
| Exploit ScenarioDerived | An authenticated low-privilege user with query access sends crafted polymorphic join 'where' filters against a collection, using equality/comparison operators to blind-infer the values of hidden or read-restricted fields (such as another user's password-reset token) that are not normally returned. No public exploit code was identified at time of analysis, but the fixing commit and its regression tests effectively document the exact malicious query shape. Because the CVSS vector is PR:L, this requires an existing authenticated account and is not anonymous. |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv2 – Oct 06, 2026
Authenticated users with query access to a Payload CMS collection can infer hidden or read-restricted field values - including password-reset tokens - by abusing polymorphic join 'where' filters that were not validated against unknown field paths. The flaw affects the npm package 'payload' in the ranges >=3.0.0 <3.90.0 and >=4.0.0-canary.0 <4.0.0-canary.34, requires a valid account holding query access (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:H/I:N/A:N), and is a moderate-priority confidentiality-only issue (CWE-200); there is no public exploit identified at time of analysis, and vendor patches exist in 3.90.0 and 4.0.0-canary.34.
Earlier versions of this analysis (1)
Information disclosure in Payload CMS npm packages (>= 3.0.0 and < 3.90.0, plus >= 4.0.0-canary.0 and < 4.0.0-canary.34) allows a user with query access to infer the values of hidden or read-restricted fields - including password-reset tokens - by abusing filter conditions on polymorphic joins. Exploitation is indirect and oracle-based rather than a direct field read: an attacker submits join 'where' filters (via REST joins[...].where or the Local API payload.find) against collections containing hidden (hidden: true) or access-restricted fields and observes which filters match. The assessed vector is CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:N/A:N, indicating a network-reachable, authenticated low-privilege attacker with high attack complexity and high confidentiality impact. No public exploit code and no confirmed active exploitation were identified at time of analysis.
RemediationAI
Vendor-released patch: upgrade Payload packages to 3.90.0 or later for the 3.x line, or to 4.0.0-canary.34 or later for the 4.x canary line - this is the only complete fix, since the upstream advisory states no full workaround exists. Where immediate upgrade is not possible, reduce (but do not eliminate) exposure by removing read access for untrusted roles on collections containing sensitive fields such as password-reset tokens, and by auditing the schema for polymorphic joins that point at collections with hidden:true fields; note that the advisory explicitly warns these controls are compensating-only and leave residual risk because the inference oracle may still respond for fields the querier is otherwise permitted to reach. The validating commit caa3f69b5aecc6e7c2b459ed87972cbca9f33ee5 confirms the fix rejects unknown polymorphic join paths with HTTP 400 in REST and a 'path cannot be queried' error in the Local API, so verifying those behaviors post-upgrade is a useful regression check. Consult the advisory at https://github.com/payloadcms/payload/security/advisories/GHSA-fpww-c55p-cjv6 and the release notes at https://github.com/payloadcms/payload/releases/tag/v3.90.0 for the authoritative upgrade guidance, and rotate any password-reset tokens or other secrets that the affected fields may have exposed.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 2 packages in npm, euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
Payload CMS is a Node.js/TypeScript headless content management framework that exposes collections through a REST and Local API. It supports 'join' fields, which let a query traverse relationships from one collection into another using nested 'where' clauses; polymorphic joins are a variant where the target collection can vary by row. The root cause (CWE-200, Exposure of Sensitive Information to an Unauthorized Actor) lives in packages/payload/src/database/queryValidation/validateSearchParams.ts: the validator previously treated any schema-safe field path inside a polymorphic join as acceptable whenever a polymorphicJoin was present, without confirming that the referenced field actually existed on the resolved target collection. Because an attacker-supplied where clause such as { hiddenSecret: { equals: 'secret' } } was still executed against the underlying data store, the presence or absence of matching results acted as an oracle for values of fields marked hidden:true or protected by read access control. The remediation commit (caa3f69b5aecc6e7c2b459ed87972cbca9f33ee5) adds an isUnknownPolymorphicJoinField check so that a polymorphic join path with no corresponding field definition is rejected rather than queried, and the accompanying tests assert HTTP 400 / thrown 'The following path cannot be queried' errors for both REST and Local API usage.
Know when the next one hits your stack
- Watch products and get an email when something new is published
- Check this CVE against your own stack and servers
- Exploit and PoC references, threat intelligence from 22 sources
No credit card · 7-day full trial
Severity by source
CVSS:4.0/AV:N/AC:L/AT:N/PR:L/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N
Exploited over the network query API by an authenticated low-privilege user (PR:L), no user interaction, leaking confidential field values only, so C:H with I:N/A:N and unchanged scope.
Primary rating from Vendor (https://github.com/payloadcms/payload).
CVSS VectorVendor: https://github.com/payloadcms/payload
Lifecycle Timeline
3DescriptionCVE.org
Impact
A user with query access could use polymorphic join filters to infer hidden or read-restricted field values, including password-reset tokens.
You are affected if:
- You use an affected Payload version.
- Users can query a collection with a polymorphic join to sensitive fields.
Patches
Users should upgrade Payload packages to >= 3.90.0 or >= 4.0.0-canary.34.
Workarounds
There is no complete workaround. Restricting read access to sensitive collections reduces exposure but does not replace upgrading.
Technical POC Analysis
Advisory: GHSA-fpww-c55p-cjv6.
Full technical POC analysis - root cause, step-by-step exploitation and patch breakdown - is part of the Pro and Business plans.
Latest POC
1 PoC candidate from github (unverified)
Sign in to open the public PoC links collected for this CVE.
Executive ViewAI
Payload CMS, a content management system used by organizations to manage digital content, contains a high-severity vulnerability in versions 3.0.0 through 3.89.x and 4.0.0-canary.0 through 4.0.0-canary.33 that allows authenticated users with query permissions to access hidden field values, including sensitive password-reset tokens. This could enable an attacker with a low-privileged account to compromise other user accounts and access restricted data, potentially leading to broader system compromise. Leadership should prioritize patching because the vulnerability is easy to exploit by any authenticated user and affects confidentiality of critical data; no public exploit identified at time of analysis.
Apply vendor patch.
Threat Intelligence Overview
How this decision was reached
The engine reads each signal below and applies a fixed list of rules from top to bottom. The first rule that matches sets the recommended action – like a statement, every line is shown.
What every action level means
SSVC AssessmentENISA
Priority Score
Attack Technique
Information disclosure occurs when an application unintentionally exposes sensitive data that aids attackers in reconnaissance or directly compromises security.
EUVD-2026-93489
GHSA-fpww-c55p-cjv6
NVD
CVE.org
EPSS
Vulnrichment
Exploit-DB