Description

In JetBrains IntelliJ IDEA before 2026.3 code execution via the document import was possible in untrusted projects

Executive Summary

Technical Details

Mitigations

References

Additional Metadata

{
  "score": "6.3",
  "pubDate": "2026-11-06T11:17:09.860Z",
  "pubdate": "2026-11-06T11:17:09.860Z",
  "executiveSummary": "A critical security vulnerability exists in JetBrains IntelliJ IDEA versions prior to 2026.3, allowing for arbitrary code execution through the document import mechanism.\nThe vulnerability is classified as an improper input validation or insecure deserialization flaw during the parsing of imported project documents.\nAn unauthenticated attacker who manages to induce a user into opening a maliciously crafted project or importing a tainted document can achieve remote code execution (RCE) within the context of the application.\nThis vulnerability poses a significant risk to the integrity, confidentiality, and availability of the host system, as successful exploitation grants the attacker the ability to execute arbitrary commands with the privileges of the IDE process.\nThe attack is primarily facilitated via untrusted projects, necessitating strict user discretion when handling external or untrusted source repositories.\nNo specific authentication or network exposure is required if the victim executes the malicious payload locally.",
  "technicalDetails": "The vulnerability resides within the document import subsystem of IntelliJ IDEA, which fails to adequately sanitize or validate input structures during the ingestion process. In versions prior to 2026.3, the import logic processes external files without sufficient isolation or verification of the data contents, allowing for the injection of malicious instruction sets.\nThe root cause is identified as a flaw in the handling of project metadata or document definitions that are processed when an untrusted project is loaded. When the IDE parses these imported documents, it fails to constrain the execution context, allowing embedded code or configuration-based triggers to interact with the underlying host system via the application's runtime.\nThe attack flow proceeds as follows: First, an attacker constructs a malicious document or a project configuration file containing obfuscated payloads that exploit the document import parser. Second, the target user opens the untrusted project or triggers an import operation within the IDE. Third, the vulnerability is triggered during the parsing phase, where the IDE interprets the malicious data as legitimate commands. Finally, the payload executes within the application process space, granting the attacker arbitrary code execution capabilities.\nBecause the IDE often operates with the user's full privileges, the post-exploitation impact includes unauthorized data exfiltration, local system manipulation, and potential lateral movement within the network. The exploitation is silent, meaning the user may not receive immediate feedback that their system has been compromised upon opening the project.\nThis vulnerability is particularly dangerous because it bypasses typical security perimeters by leveraging a core functionality—document import—that users rely on for daily development tasks. No specific privilege level beyond standard user rights is required to trigger the vulnerability, and the complexity of modern IDE projects makes it difficult for users to inspect the underlying configuration files for malicious code before import."
}