Description

IBM Langflow OSS 1.0.0 through 1.9.6 could allow a remote attacker to obtain unauthorized access to user accounts due to improper restriction of excessive authentication attempts.

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "9.1",
  "pubDate": "2026-08-13T21:17:45.870Z",
  "pubdate": "2026-08-13T21:17:45.870Z",
  "executiveSummary": "A vulnerability affecting IBM Langflow OSS versions 1.0.0 through 1.9.6 permits remote attackers to secure unauthorized access to user accounts. Classified as an improper restriction of excessive authentication attempts, this security flaw exposes authentication interfaces to automated brute-force attacks and credential stuffing vectors without adequate rate-limiting or account lockout mechanisms.\nThe primary impact of this vulnerability involves the compromise of user accounts and sensitive operational data stored within the platform. Remote unauthenticated actors can leverage automated tooling to systematically query authentication endpoints, iterating through potential credential combinations until successful access is achieved.\nThe risk implications are significant for deployments handling sensitive workflows, as successful exploitation bypasses standard authentication boundaries, potentially leading to unauthorized execution of flows, data exfiltration, and complete administrative takeover depending on the compromised account privileges.\nExploitation requirements are minimal, necessitating only network connectivity to the exposed authentication interface and a malicious script or tool capable of executing high-frequency authentication requests. No prior privileges or specialized authentication tokens are required to initiate the attack flow.",
  "technicalDetails": "The root cause of this vulnerability lies in the insufficient enforcement of rate-limiting controls and the absence of a robust throttling or locking mechanism for repeated authentication attempts within the authentication component of IBM Langflow OSS versions 1.0.0 through 1.9.6.\nFrom a network exposure perspective, the vulnerable authentication endpoints are accessible remotely, allowing any unauthenticated external entity to interact directly with the login functionality over standard network protocols without artificial delays or challenge-response roadblocks.\nThe exploitation method relies on brute-force or credential-stuffing techniques. An attacker initiates the attack flow by dispatching a high volume of concurrent or sequential HTTP authentication requests targeting the application login interface.\nBecause the affected component fails to track, limit, or penalize excessive authentication failures originating from a single source or targeting a specific user account, the application processes every credential submission iteratively.\nThe payload behavior involves automated request generation using dictionaries of compromised credentials or common passwords. The system processes these requests normally, returning distinguishing response codes, headers, or payload lengths for successful versus failed authentication attempts, which the attacker parses to identify valid credentials.\nPost-exploitation impact includes full session establishment under the identity of the target user. Once unauthorized access is attained, the malicious actor inherits the privileges associated with the compromised account, enabling them to inspect, modify, or execute data flows, interact with integrated backend services, and potentially pivot to broader infrastructure assets connected to the Langflow instance."
}