Attack ChainAIDerived
Hypothetical attack flow derived from CVE metadata – not a confirmed exploit chain
-
AccessAuthenticate as low-privilege user
-
ExploitInvoke AI provider configuration handler
-
ExecutionTrigger cross-session data mix-up in ai_client.py
-
ImpactRead another session's leaked configuration data
Vulnerability Assessment
| Weakness |
CWE
ASVS 5.0
V14 - Data Protection Exposure of data element to wrong session
|
| Exploitation Conditions | Requires a valid authenticated low-privilege account on an OpenConstructionERP instance at version ≤14.8.1 (PR:L); exploitation targets the AI Provider Configuration Handler implemented in backend/app/modules/ai/ai_client.py. No user interaction is needed (UI:N) and the attack is remote over the network (AV:N/AC:L). The limiting factor is the authentication requirement plus the need for concurrent/overlapping sessions for the wrong-session data element to be exposed; impact is confined to data accessible through the AI configuration handler rather than full system compromise. Fixed in v15.0.0 (pip/PyPI distribution; note the v15.0.0 desktop installer is broken per vendor, use pip or await v15.0.1). |
| Impact | All signals point to a low-to-moderate, not a priority, issue. The CVSS 4.0 base vector (AV:N/AC:L/AT:N/PR:L/UI:N/VC:L/VI:L/VA:L) scores 5.3 and already bakes in E:P (proof-of-concept exploit maturity), so the published 5.3 reflects a PoC-grade, authenticated, low-impact flaw rather than a wormable one. Exploitation requires PR:L - an authenticated low-privilege user - which materially caps blast radius. POC is publicly available (GitHub security advisory GHSA-wfpw-cv5v-64j5), so publicly available exploit code exists, but this is distinct from active exploitation: there is no CISA KEV listing and no EPSS score supplied, so there is no evidence of in-the-wild use. CWE-488 (exposure of data element to wrong session) typically manifests as a cross-tenant/cross-session leak of AI provider configuration data; the impact is primarily confidentiality with limited integrity, consistent with the VC:L/VI:L rating. Net: a real but low-severity authenticated information-exposure bug with public PoC, worth patching on normal cadence rather than emergency response. Note a reporting caveat - this is a VulDB-sourced CVE (vuln 412354) with vendor-confirmed GitHub advisory and patch, lending the core facts good provenance. |
| Exploit ScenarioDerived | An authenticated low-privilege user of an OpenConstructionERP instance (≤14.8.1) interacts with the AI Provider Configuration Handler, and due to improper session isolation in backend/app/modules/ai/ai_client.py receives data element(s) belonging to another user's session - leaking AI provider configuration (potentially API keys/endpoints) across sessions. Public exploit code exists in the vendor's GitHub security advisory (GHSA-wfpw-cv5v-64j5), lowering the skill required, and the attack is remote (AV:N) but gated by needing a valid low-privilege account (PR:L). |
Get an e-mail when this CVE is exploited, patched or re-scored. Free account.
Alert meAnalysisAIv1 – Oct 01, 2026
OpenConstructionERP through 14.8.1 can leak one user's data to another user's session through the AI Provider Configuration Handler implemented in backend/app/modules/ai/ai_client.py. The issue is only reachable by an account that is already authenticated on the instance (CVSS vector PR:L) and impact is limited to confidentiality and integrity of AI provider configuration data rather than full system compromise; the independent assessment rates the overall risk as low-to-moderate rather than a priority issue. Publicly available exploit code exists and the CVSS 4.0 vector carries an E:P (proof-of-concept) exploit-maturity score, so the flaw should still be treated as practically exploitable, but there is no confirmed actively exploited (CISA KEV) status for this CVE.
RemediationAI
Upgrade to OpenConstructionERP v15.0.0 or later, which the vendor states is sufficient to resolve the issue (CVSS 4.0 fix; vendor advisory https://github.com/datadrivenconstruction/OpenConstructionERP/security/advisories/GHSA-wfpw-cv5v-64j5, release https://github.com/datadrivenconstruction/OpenConstructionERP/releases/tag/v15.0.0). Be aware that the v15.0.0 desktop installers for Windows, macOS, and Linux do not start because the sidecar was built without its translation catalogue; the vendor's release notes state the PyPI/pip package is unaffected, so installing or upgrading via pip from PyPI is the practical path, otherwise stay on 14.8.1 and wait for v15.0.1, which the vendor says carries the build fix and a corrected release. If upgrading is not immediately possible, compensating controls are: remove or disable the AI provider configuration feature for untrusted accounts and restrict the AI configuration endpoints to a small set of administrators (this reduces the exposed attack surface but removes an operational feature for regular users); rotate any AI provider API keys stored in the configuration, since a wrong-session disclosure may have exposed credentials (key rotation invalidates existing integrations until they are reconfigured); and avoid concurrent multi-user sessions against the same instance until the upgrade is applied, which limits exposure to the overlapping-session precondition but is not practical for normal production use.
Affected VersionsENISA
Affected Packages
Source: OSV (Open Source Vulnerabilities)
Affects your stack
Affects 1 package in euvd
Checking this CVE against your own stacks and servers is part of the Basic plan and up.
Technical ContextAI · show details
The root cause class is CWE-488, Exposure of Data Element to Wrong Session, a session-management flaw in which state belonging to one request or user is written to a context that another concurrent session can read. The affected component is the AI Provider Configuration Handler inside the backend Python module backend/app/modules/ai/ai_client.py, which manages settings for outbound AI/LLM provider connections such as endpoint, model, and API credentials. The CPE entry cpe:2.3:a:datadrivenconstruction:openconstructionerp:*:*:*:*:*:*:*:* identifies the vendor and product as datadrivenconstruction OpenConstructionERP across the affected range, and the vulnerability maps to the vendor advisory GHSA-wfpw-cv5v-64j5 in the project's GitHub repository. Because the flaw concerns session-scoped data being shared or cached at a module scope reachable by the AI client, exploitation generally requires overlapping or concurrent sessions on a reachable instance, which explains the S:U scope and the low confidentiality/integrity scores in the authoritative assessment (CVSS:3.1/AV:N/AC:L/PR:L/UI:N/S:U/C:L/I:L/A:N).
Know when Openconstructionerp gets its next vulnerability
- 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:L/VI:L/VA:L/SC:N/SI:N/SA:N/E:P
Remote but requires a low-privilege authenticated account (PR:L); cross-session exposure yields limited confidentiality and integrity impact, no availability impact.
Primary rating from Vendor (VulDB).
CVSS VectorVendor: VulDB
Lifecycle Timeline
3DescriptionCVE.org
A vulnerability was found in datadrivenconstruction OpenConstructionERP up to 14.8.1. The impacted element is an unknown function of the file backend/app/modules/ai/ai_client.py of the component Al Provider Configuration Handler. Performing a manipulation results in exposure of data element to wrong session. The attack may be initiated remotely. The exploit has been made public and could be used. Upgrading to version 15.0.0 is sufficient to resolve this issue. It is suggested to upgrade the affected component.
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-2026-90521
GHSA-8rwg-xf8h-29m3
NVD
CVE.org
EPSS
Vulnrichment
VulDB
Exploit-DB