Description

IBM Langflow OSS 1.0.0 through 1.10.2 could allow a remote authenticated attacker to change the password of an account due to improper authentication.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "5.4",
  "pubDate": "2026-09-14T20:16:42.140Z",
  "pubdate": "2026-09-14T20:16:42.140Z",
  "executiveSummary": "A critical security vulnerability has been identified in IBM Langflow OSS, specifically affecting versions 1.0.0 through 1.10.2. The flaw is categorized as an improper authentication issue, which allows a remote, authenticated attacker to manipulate user account credentials. By exploiting this vulnerability, an unauthorized party can force a password reset or change for any account, including those with administrative privileges. This represents a significant risk to the integrity and confidentiality of the application, as successful exploitation facilitates full account takeover. The attack vector is remotely exploitable, requiring only a valid authenticated session to initiate the malicious request. Given that the system's security architecture fails to verify the authorization context of the password change request, the impact is severe, potentially leading to unauthorized data access, modification of workflow configurations, and complete administrative compromise of the Langflow environment.",
  "technicalDetails": "The vulnerability resides in the authentication logic implemented within IBM Langflow OSS versions 1.0.0 to 1.10.2. The root cause is a failure in the application's backend to adequately validate the identity and authorization context during the password modification lifecycle. Specifically, the API endpoint responsible for managing user account settings and password updates does not enforce mandatory checks that verify whether the requester has the authority to modify the target account's credentials.\nExploitation is achieved through a remote attack vector where an attacker, having already authenticated to the platform with a standard or low-privileged user account, crafts a specially formed HTTP request targeting the password update function. Because the server-side code relies on client-provided parameters without sufficient server-side validation against the current session's user identity, the attacker can supply the target username or user ID within the request payload. The application incorrectly trusts this input, leading it to update the password of the specified account in the underlying database or user management store.\nThe attack flow follows these steps: 1) The attacker establishes a legitimate authenticated session within the IBM Langflow environment. 2) The attacker identifies the API endpoint or underlying function responsible for the password change process via reconnaissance of the network traffic or application source code. 3) The attacker initiates an HTTP request to this endpoint, systematically altering the payload to include the identifier of the victim account. 4) The server processes the request, bypasses the expected ownership validation check, and executes a database update command to override the victim's credentials with the attacker-supplied value. 5) Upon completion, the attacker can leverage the new credentials to gain unauthorized access to the target account. This exploitation path highlights a significant flaw in session-to-action binding and privilege enforcement. The post-exploitation impact includes full account takeover, which in the context of Langflow, allows the attacker to view, execute, and modify sensitive workflows, internal data integrations, and potentially pivot to underlying system resources if the account possesses elevated permissions. The lack of granular server-side authorization enforcement remains the primary vector for this vulnerability."
}