The vulnerability lies in the create_project/encrypt_auth_settings function of the Project Creation Endpoint, which writes authentication settings directly to a file on disk in cleartext. This results in cleartext storage of potentially sensitive credentials, exposing them to anyone with read access to the file system. The weakness is identified as CWE‑312 (Cleartext Storage of Sensitive Information) and CWE‑313 (Insecure Encryption). An attacker who obtains the authentication settings can gain unauthorized access to services the credentials protect, thereby compromising confidentiality and possibly enabling further attacks such as privilege escalation. The affected product is langflow‑ai's Langflow platform, with versions up to and including 1.8.3. The flaw resides specifically in src/backend/base/Langflow/api/v1/projects.py and impacts the Project Creation Endpoint provided by the API. The risk is moderate, reflected by a CVSS score of 5.3. No EPSS score is available, and the vulnerability is not listed in the CISA KEV catalog. The attack can be launched remotely against the API, and the exploit has been publicly disclosed. While no automated exploitation package is reported, the cleartext storage provides a straightforward method for an attacker to retrieve credentials once they have any level of access to the system’s filesystem or the running application’s environment.

The vulnerability affects the langflow‑ai Langflow application versions up to 1.8.3. This includes the Project Creation Endpoint that writes authentication settings to disk in cleartext.

The CVSS score of 5.3 indicates a medium severity. The EPSS score is not available, and the vulnerability is not in the CISA KEV list. Attackers can exploit the flaw remotely by triggering the Project Creation Endpoint with manipulated auth_settings, causing the application to persist sensitive data in cleartext. Because the flaw lies in the backend code and the attack surface is the exposed API, the likelihood of exploitation depends on the exposure of the API to untrusted users and the presence of any authentication controls around it.