CVE-2026-0768
Description
Langflow code Code Injection Remote Code Execution Vulnerability. This vulnerability allows remote attackers to execute arbitrary code on affected installations of Langflow. Authentication is not required to exploit this vulnerability. The specific flaw exists within the handling of the code parameter provided to the validate endpoint. The issue results from the lack of proper validation of a user-supplied string before using it to execute Python code. An attacker can leverage this vulnerability to execute code in the context of root. . Was ZDI-CAN-27322.
In plain language
AI Act nowCVE-2026-0768 is a remote code execution flaw in Langflow that can let an attacker run arbitrary code without logging in; you should treat it as urgent if your Langflow “validate” endpoint is reachable from the network.
CVE-2026-0768 is a unauthenticated remote code execution (CWE-94) in Langflow’s handling of the user-supplied code parameter to the validate endpoint, where insufficient input validation lets attackers inject and execute Python code on the server.
What to do now
- Check whether your Langflow instance is reachable from the internet (or any untrusted network) and whether the validate endpoint is exposed.
- If exposed, immediately restrict access to the validate endpoint to trusted IPs only (or block it entirely at the firewall/reverse proxy).
- Review your current Langflow version and look for an official security update for CVE-2026-0768; if none is available yet, pause public access until a fix is released.
- Hunt for signs of abuse in Langflow logs and web/proxy logs around recent requests to the validate endpoint (especially repeated validate calls from unexpected sources).
- Apply the vendor’s fixed version as soon as it is published for your branch, and remove the temporary exposure restrictions only after the fix is in place.
CVSS Vector Breakdown
AV:NAttack VectorAC:LAttack ComplexityPR:NPrivileges RequiredUI:NUser InteractionS:UScopeC:HConfidentialityI:HIntegrityA:HAvailabilityWeaknesses
Affected Products
Exploitability
Attack Graph
Click technique nodes for MITRE ATT&CK details · drag to pan · Ctrl/⌘ + scroll to zoom, or go fullscreen.
MITRE ATT&CK
2 techniquesReferences
- Critical Langflow Flaw Exploited as Attacks on AI Platform Riseen·Dark Reading· Exploited Langflow rce
- Critical Langflow flaw exploited to steal OpenAI and AWS keysen-us·BleepingComputer· Exploited Langflow rce
- Hackers Start Exploiting Critical Langflow Vulnerabilityen-us·SecurityWeek· Exploited Langflow rce
- Attackers Exploit Critical Langflow and Rails Flaws in Credential-Probing and C2 Activityen·The Hacker News· Exploited Langflow rce
Unlock Complete Vulnerability Intelligence
Get the full picture for CVE-2026-0768 and every CVE in our database. Create a free account — no credit card required.
Create Free Account