Langflow: attackers can bypass security checks without logging in
Langflow is a tool for building AI workflows. A flaw lets someone outside your network skip the login checks and get access they should not have.
| Severity | CriticalCVSS 9.8 of 10 |
|---|---|
| Being exploited? | UnclearNo reports either way yet |
| Fix available? | UnclearNo fixed version published in the sources yet |
| How sure is this? | Official, 3 sourcesBacked by an official source |
What happened
IBM Langflow OSS versions 1.0.0 through 1.12.8 have an improper authentication flaw that allows a remote attacker to bypass security restrictions. CERT-Bund reports several related Langflow vulnerabilities that could allow code execution, denial of service, information disclosure, file manipulation and security control bypass.
Are you affected?
You are affected if you run IBM Langflow OSS version 1.0.0 through 1.12.8 and it is reachable by untrusted users or the internet.
IBM Langflow OSS 1.0.0 through 1.12.8
What to do
- Check whether your Langflow version is between 1.0.0 and 1.12.8
- Update Langflow to a fixed version as published in the official IBM or GitHub advisory
- Restrict network access to Langflow until the update is applied
- Review logs for unexpected access to Langflow
Timeline
- 29 Oct 2026, 16:20 UTCPublicly disclosed
- 29 Oct 2026, 16:20 UTCReported by NVD · NVD
- 29 Oct 2026, 16:20 UTCPicked up by Zero Radar
- 29 Oct 2026, 20:34 UTCGitHub Advisory reported on it · GitHub Advisory
- 30 Oct 2026, 10:50 UTCCERT-Bund (BSI) reported on it · CERT-Bund (BSI)
Sources 3 reports, oldest first
Official vendor, government CERT or advisory database
- CVE-2026-93680NVD · 29 Oct 2026, 16:20 UTC
- IBM Langflow OSS 1.0.0 through 1.12.8 could allow a remote attacker to bypass security...GitHub Advisory · 29 Oct 2026, 18:31 UTC
- [UPDATE] [hoch] Langflow: Mehrere SchwachstellenCERT-Bund (BSI) · 30 Oct 2026, 10:49 UTC
Technical details
- ID
- CVE-2026-93680 · GHSA-vfwm-g9g9-v2g5
- CVSS
- 9.8 / 10
- EPSS
- 2.2% chance of exploitation within 30 days
- Disclosed
- 29 Oct 2026, 16:20 UTC
- On Zero Radar
- 29 Oct 2026, 16:20 UTC (0 min after disclosure)
- Last change
- 30 Oct 2026, 10:50 UTC
The plain-language parts and the attack diagram were written by an AI model (anthropic/claude-haiku-5.5) from the sources above. Facts like versions, exploitation and dates come from the sources themselves. How Zero Radar works