Langflow OSS EUVD-2026-113158

Requires an authenticated session on the Langflow OSS instance (a valid account at low privilege, matching PR:L) plus use of the flow-export or project-export functionality; the exported artifact fails to sanitize embedded secrets, disclosing them to the exporting user. Exploitation is limited by the need for valid credentials and by the fact that it only exposes secrets already stored within accessible flows/projects - it does not grant code execution, modification, or denial of service. The precise secrets leaked and whether low-privilege users can export flows they do not own are not detailed in the available sources; updating to the IBM-patched version is the remediation the vendor provides.

This is a moderate-priority information-disclosure issue, not a drop-everything emergency, and the available signals are internally consistent. The CVSS 3.1 vector (AV:N/AC:L/PR:L/UI:N, C:H/I:N/A:N, scope unchanged) describes a network-reachable but authenticated flaw: an attacker needs valid low-privilege credentials, there is no integrity or availability impact, and the only consequence is confidentiality loss (exposed secrets). The vuln.today decision label is 'this_month', which aligns with a 6.5/Medium severity that is real but bounded by the authentication requirement. No EPSS score, no CISA KEV listing, and no PoC or exploit reference are present in the input, so there is no evidence of active or imminent mass exploitation - this looks like a hygiene/remediation item rather than a widely-weaponized threat. The main judgment call is the confidentiality impact: C:H is appropriate if the leaked secrets are high-value credentials (API keys, model-provider tokens, datastore passwords embedded in flows), which is plausible given the product's purpose, but the input does not enumerate exactly which secrets leak, so the severity of downstream compromise depends on what the victim stored in their flows. No signals conflict; the chief limiting factor is that exploitation requires an authenticated account.

A remote attacker who already holds a valid low-privilege Langflow OSS account uses the built-in flow or project export feature; because secrets are not properly sanitized before export, the resulting export artifact contains credentials (such as API keys or connection secrets) the user embedded in their flows. The attacker reads these secrets from the export and can reuse them against the referenced external services. No public exploit identified at time of analysis.