JetBrains Rider’s AI Assistant can auto‑update third‑party skills without user approval, potentially installing malicious code. This vulnerability exposes the development environment to unauthorized code execution, threatening confidentiality, integrity, or availability of the user session and projects that rely on the skill.

The issue affects JetBrains Rider before version 2026.2.1. No additional version matrix is provided, so all earlier releases are vulnerable unless the user has disabled auto‑updates through the IDE settings.

The CVSS score of 4.8 indicates moderate severity. EPSS is not available and the vulnerability is not listed in CISA KEV, suggesting no known public exploits. Based on the description, it is inferred that the attack vector involves an attacker supplying a malicious third‑party skill that the AI Assistant will install automatically; because the update occurs without user confirmation, the code could run with the user’s credentials. This risk is primarily relevant for environments that enable third‑party skill installation and may lead to local privilege escalation or code execution within the developer’s workspace.