Langflow OSS: Code execution via stored Python components in public workflows
Summary
IBM Langflow OSS versions 1.0.0 through 1.12.4 contain a code injection flaw (CWE-94). A remote attacker can trigger Python component code already stored in a public workflow. The code then runs under the flow owner's runtime context inside the Langflow server process, bypassing the rule that public execution paths may only run server-trusted components. The CVE data provided lists CVSS 9.8, while IBM's bulletin rates the issue 6.5 with a different vector, and the authentication requirement is not confirmed, so check the final score in NVD.
Affected
- IBM Langflow OSS 1.0.0 through 1.12.4
- Desktop builds may use different patch numbers, but major and minor versions align.
Fixed in
- Fixed in IBM Langflow OSS 1.11.2 according to the IBM CVE record.
- IBM recommends upgrading to Langflow OSS 1.12.5.
Workaround
- No CVE-specific workaround published by IBM.
- Keep LANGFLOW_ALLOW_PUBLIC_CUSTOM_COMPONENTS disabled, which is the default for public workflows.
- For deployments using MCP stdio, IBM recommends setting LANGFLOW_MCP_SERVER_INTERPRETER_HARDENING=true and LANGFLOW_CUSTOM_COMPONENT_ADMIN_ONLY=true, then restarting the service. These controls do not replace the upgrade.
- Unpublish public flows that are not needed, and place remaining instances behind an authenticated reverse proxy or VPN.
Exploitation
- CISA SSVC: exploitation none, automatable yes, technical impact partial.
- OffSeq reports no known exploitation as of 29 Oct 2026. VulDB reports no exploit available at publication.
- No public proof of concept found.
Our recommendation
- Inventory all Langflow OSS and Desktop instances and record their exact versions.
- Upgrade to Langflow OSS 1.12.5 and test your flows, following IBM's upgrade instructions and backing up first.
- Review public flows, especially those containing Python components, and unpublish those not required.
- Remove unnecessary internet exposure of Langflow instances.
- Monitor for unexpected shell or interpreter processes started by the Langflow process.
- Verify the final CVSS score and authentication requirement in NVD and the IBM bulletin.
Public exploits
PoC-in-GitHub, Metasploit and Exploit-DB list nothing for this CVE.
Research sources
- IBM Security Bulletin: Langflow OSS is affected by multiple vulnerabilities www.ibm.com
- Langflow release notes docs.langflow.org
- VulDB: CVE-2026-18894 vuldb.com
- OffSeq Radar: CVE-2026-18894 radar.offseq.com
- Strix: CVE-2026-18894 www.strix.ai
Researched and written up in English and German by an AI (Claude by Anthropic) based on NVD, CISA, EPSS and the sources listed. The vendor information is authoritative. Researched on 30 Oct 2026, 16:05.
Original description (NVD)
IBM Langflow OSS 1.0.0 through 1.12.4 could allow a remote attacker to trigger pre-existing stored Python component code in a public workflow, executing it under the flow owner's runtime context within the server process. This bypasses the intended trust boundary where public execution paths are restricted to server-trusted component code.
NVD references (1)
Sources
The tracker combines official, freely usable data sources – updated continuously:
- NVD (NIST) – CVEs, rating (CVSS), affected products
- CISA KEV – vulnerabilities confirmed by CISA as actively exploited
- CISA Vulnrichment – CISA assessment (SSVC) in the official CVE record
- FIRST EPSS – probability of exploitation
- Microsoft MSRC – Patch Tuesday: vulnerabilities, severity, updates
- PoC-in-GitHub – public proof-of-concept repositories
- Metasploit (Rapid7) – modules of the Metasploit Framework
- Exploit-DB (OffSec) – archive of public exploits (links only)
- Patch days: official vendor pages (linked per vendor) (Vendor patch days)
This product uses data from the NVD API but is not endorsed or certified by the NVD. EPSS data © FIRST.org.
Affected? We can help.
We check whether your systems are affected, install updates and close the gap – continuously with the System Health Checker if you wish.