Is CVE-2025-68478 exploited in the wild?

No. Junglewise Threat Intelligence has no report of CVE-2025-68478 being exploited in the wild.

How severe is CVE-2025-68478?

High severity, with a CVSS score of 7.1, published 19 December 2025.

Which products and versions are affected?

Langflow < 1.7.1; langflow (PyPI) ECOSYSTEM:introduced=0,fixed=1.7.0, ECOSYSTEM:introduced=0,fixed=1.7.1.

Is there a fix for CVE-2025-68478?

Yes. langflow (PyPI): upgrade to 1.7.0, 1.7.1 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

How is CVE-2025-68478 attacked?

It is attacked over the network, in a software library.

Am I affected by CVE-2025-68478?

Check whether you run Langflow or langflow (PyPI) 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.

CVE-2025-68478: Langflow arbitrary file write via fs_path parameter

Q: Which products and versions are affected?

Langflow < 1.7.1; langflow (PyPI) ECOSYSTEM:introduced=0,fixed=1.7.0, ECOSYSTEM:introduced=0,fixed=1.7.1.

Q: Is there a fix for CVE-2025-68478?

Yes. langflow (PyPI): upgrade to 1.7.0, 1.7.1 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.

Q: How is CVE-2025-68478 attacked?

It is attacked over the network, in a software library.

Q: Am I affected by CVE-2025-68478?

Check whether you run Langflow or langflow (PyPI) 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.

CVE-2025-68478 · Severity: high · CVSS 7.1 · Published 2025-12-19

Langflow, a framework for building multi-agent AI applications, contains a vulnerability that allows authenticated users to write files to arbitrary locations on the server. By providing a specific file path when creating a 'Flow,' an attacker can overwrite critical system or application files with JSON data. This could lead to service disruptions, corruption of configuration files, or unauthorized modification of application data.

Technologies: Langflow, langflow (PyPI). Vendors: Langflow, PyPI.

Quick answers

Is CVE-2025-68478 exploited in the wild?: No. Junglewise Threat Intelligence has no report of CVE-2025-68478 being exploited in the wild.
How severe is CVE-2025-68478?: High severity, with a CVSS score of 7.1, published 19 December 2025.
Which products and versions are affected?: Langflow < 1.7.1; langflow (PyPI) ECOSYSTEM:introduced=0,fixed=1.7.0, ECOSYSTEM:introduced=0,fixed=1.7.1.
Is there a fix for CVE-2025-68478?: Yes. langflow (PyPI): upgrade to 1.7.0, 1.7.1 or later. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
How is CVE-2025-68478 attacked?: It is attacked over the network, in a software library.
Am I affected by CVE-2025-68478?: Check whether you run Langflow or langflow (PyPI) 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.

How to fix

langflow (PyPI): upgrade to 1.7.0, 1.7.1 or later
```
pip install "langflow>=1.7.1"
```

Technical details

An arbitrary file write vulnerability exists in Langflow's flow creation endpoint due to insufficient validation of the 'fs_path' parameter. When a user creates a flow, the backend uses the user-provided path in the '_save_flow_to_fs' function without normalization or directory restriction. An authenticated attacker with a valid API key or JWT can provide absolute or relative paths (e.g., /tmp/POC.txt) to create or overwrite files with the serialized JSON representation of the Flow object. While the content is restricted to JSON, the ability to overwrite arbitrary files within the application's process permissions can lead to integrity loss or denial of service. This issue is addressed in version 1.7.1.

Affected products

langflow-ai langflow < 1.7.1

Identifiers

CVE-2025-68478
GHSA-f43r-cc68-gpx4
PYSEC-2025-125

Timeline

2025-12-19: disclosed
2025-12-19: advisory
1.7.1: patched

References

This page as Markdown. Free to reuse under CC BY 4.0 with a link back.