The flaw allows an attacker to embed malicious script in a Jupyter notebook markdown cell that is opened in JetBrains PyCharm, causing the script to run in the IDE’s renderer. The consequence is the ability to execute arbitrary code within the IDE, potentially steal credentials or modify local files. This is an example of the input‑sanitization weakness classified as CWE‑79.

JetBrains PyCharm versions prior to 2025.3.4 are affected. The vulnerability applies to all installations that use the built‑in Jupyter notebook support, regardless of operating system.

The CVSS score of 6.1 indicates moderate severity. EPSS data is not available and the vulnerability is not listed in the KEV catalog, suggesting no known exploitation activity. Exploitation requires the attacker to supply a malicious notebook file that the user opens in PyCharm, so user interaction is a prerequisite. Users who open notebooks from untrusted or unknown sources are at risk of having arbitrary scripts executed in the IDE environment.