# CVE-2026-3357: IBM Langflow Desktop remote code execution in FAISS component

Source: https://junglewise.ai/threats/cve-2026-3357-ibm-langflow-desktop-remote-code-execution-in-faiss-component  
Publisher: Junglewise Threat Intelligence, https://junglewise.ai/threats  
Published: 2026-04-08. Updated: 2026-07-25  
License: CC BY 4.0 with a link back

CVE-2026-3357 · Severity: high · CVSS 8.8 · Published 2026-04-08

IBM Langflow Desktop, a tool used for building AI and machine learning workflows, contains a security flaw in its data processing component. An authorized user could exploit this flaw to run unauthorized commands on the underlying system by uploading a malicious file. This could lead to a full system takeover, data theft, or disruption of AI operations.

Technologies: [IBM Langflow Desktop](https://junglewise.ai/threats/technologies/langflow-desktop), [Langflow](https://junglewise.ai/threats/technologies/langflow). Vendors: [IBM](https://junglewise.ai/threats/vendors/ibm), [Langflow](https://junglewise.ai/threats/vendors/langflow).

## Quick answers

**Is CVE-2026-3357 exploited in the wild?** No. Junglewise Threat Intelligence has no report of CVE-2026-3357 being exploited in the wild.

**How severe is CVE-2026-3357?** High severity, with a CVSS score of 8.8, published 8 April 2026.

**Which products and versions are affected?** IBM Langflow Desktop 1.6.0 - 1.8.2.

**Is there a fix for CVE-2026-3357?** A patch was published on 2026-04-08: Version 1.8.3 released to address the vulnerability Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

**How is CVE-2026-3357 attacked?** It is attacked over the network, in a network appliance.

**Am I affected by CVE-2026-3357?** Check whether you run IBM Langflow Desktop or Langflow at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.

## Technical details

A deserialization vulnerability (CWE-502) exists in IBM Langflow Desktop versions 1.6.0 through 1.8.2. The flaw is located in the FAISS Vector Store component, which is used for retrieval-augmented generation (RAG) workflows. The component is configured by default to permit the unsafe deserialization of Python Pickle files. An authenticated attacker can exploit this by uploading a specially crafted Pickle file and configuring the FAISS component to load it. Successful exploitation allows for arbitrary code execution with the privileges of the backend service. The issue is resolved in version 1.8.3.

## Affected products

- IBM Langflow Desktop 1.6.0 - 1.8.2

## Timeline

- 2026-04-02: advisory: Initial publication by IBM
- 2026-04-08: disclosed: NVD publication date
- 2026-04-08: patched: Version 1.8.3 released to address the vulnerability

## References

- [https://www.ibm.com/support/pages/node/7268428](https://www.ibm.com/support/pages/node/7268428)

## Related threats

- [Langflow SSRF in URL-taking components](https://junglewise.ai/threats/langflow-ssrf-in-url-taking-components-protection-disabled-by-default-db7cac37)
- [Langflow authorization bypass in flow API endpoints](https://junglewise.ai/threats/cve-2026-105698-langflow-is-a-tool-for-building-and-deploying-ai-powered-agents)
- [Langflow MCP stdio command injection](https://junglewise.ai/threats/cve-2026-105697-langflow-is-a-tool-for-building-and-deploying-ai-powered-agents)
- [Langflow unsafe eval() in schema.py code execution](https://junglewise.ai/threats/cve-2026-101861-langflow-1-0-16-before-1-12-0-and-0-0-94-before-1-12-0-contain)
- [IBM Langflow OSS path traversal arbitrary file write](https://junglewise.ai/threats/cve-2026-84889-ibm-langflow-oss-path-traversal-arbitrary-file-write)