Skip to main content
← All vulnerabilities
CVE-2026-107326EUVD-2026-113138GHSA-WVGW-3GW6-PMPH

Langflow OSS is affected by multiple vulnerabilities

Published
Last modified
published
Not listed in CISA KEV
Specific setups only
Fix
Fixed in 1.12.5
1 proof of concept on GitHub
  • Exploitation: none
  • Automatable: no
  • Technical impact: total

Summary

IBM Langflow OSS, an open-source tool for building AI workflows, is affected by this vulnerability. A remote attacker with a valid login could exploit it to run arbitrary commands on the underlying system.

Summarized by anthropic/claude-haiku-5.5. Verify details against the original sources.

Read the original description

IBM Langflow OSS 1.0.0 through 1.12.4 could allow a remote authenticated attacker to execute arbitrary code due to improper neutralization of special elements used in an OS command.

Timeline

  1. CVE published
  2. First seen by cve.now
  3. CISA SSVC assessment

    Exploitation: none · Automatable: no · Technical impact: total

  4. Record last updated

Affected versions

  • IBM Langflow OSS

    From 1.0.0 · through 1.12.4

References 2

ibm.com

GitHub · rmhowe425/POC-CVE-2026-107326

FAQ

Questions about CVE-2026-107326

What does this mean?

An authenticated remote user can send crafted input that gets passed into an operating system command without proper neutralization, which can let them run arbitrary code on the Langflow server.Source: ibm.com

Am I affected, and how do I check?

IBM's advisory covers Langflow OSS from 1.0.0 through 1.12.4. To check a deployment, look at the installed Langflow version, for example with the pip show langflow command, and compare it against that range.Source: ibm.com

What should I do about it?

Upgrade Langflow OSS to version 1.12.5, which IBM says fixes the issue. The MCP stdio workaround (LANGFLOW_MCP_SERVER_INTERPRETER_HARDENING=true and LANGFLOW_CUSTOM_COMPONENT_ADMIN_ONLY=true, then restart) is listed only as a temporary measure and does not replace the upgrade.Source: ibm.com

Is it being exploited?

It is not in CISA's Known Exploited Vulnerabilities catalog, and CISA's SSVC assessment does not mark exploitation as active. That does not mean it has not been attacked: vendors and researchers can report attacks before CISA lists them. Public exploit code exists (a proof of concept on GitHub).Source: cve.now feed data