Langflow Patches Two Critical Unauthenticated Remote Code Execution Flaws
Two critical flaws in Langflow, the open-source tool for building AI agents and workflows, let a remote attacker with no account bypass authentication or run arbitrary code, IBM said. The only fix is version 1.13.0; every release from 1.0.0 through 1.12.8 is affected. No exploitation has been reported and neither flaw is in CISA's KEV catalog; update now.
What happened
IBM disclosed two critical flaws in Langflow, the open-source tool for building AI agents and workflows. A remote attacker with no account can bypass authentication and run arbitrary code on the server. The fix shipped in version 1.13.0.
Why it matters
Every release from 1.0.0 through 1.12.8 is affected. A breach could give the attacker full control of the server, with risk of data theft and disruption of AI workflows. Any company running an old Langflow version is exposed.
What to do
Update Langflow to version 1.13.0, the only fix. Until then, restrict access to the tool and check servers for suspicious activity.
For the technical team
- Affected
- Langflow (plataforma open-source de agentes e fluxos de IA, com relatos de vulnerabilidade tratados pela IBM) · Langflow, all versions from 1.0.0 through 1.12.8.
- Exploitation
- Disclosed by IBM and reported by The Hacker News. No exploitation reported in the sources; neither flaw is in CISA's KEV catalog.
- Fix
- Version 1.13.0, the only available fix.
- Vector
- Two Langflow flaws allowing a remote attacker with no account to bypass authentication or run arbitrary code. Exploitation condition: remote access to the service, no valid login required.
- More details
- Langflow vulnerability reports are handled by IBM, according to The Hacker News.
Timeline
Incident opened.
