Malicious Nx npm packages reportedly weaponize AI coding agents for data exfiltration
Eight malicious versions of the Nx and Nx Powerpack npm packages were published in August 2025 and live for approximately 5 hours before removal. The malicious payload reportedly weaponized local AI coding agents to inventory and exfiltrate sensitive files, credentials, and secrets to a public GitHub repository, marking what security researchers believe may be one of the first documented cases of malware leveraging AI assistant CLIs for reconnaissance and data exfiltration.
What happened
On August 26–27, 2025, eight malicious Nx and Nx Powerpack releases were pushed to npm and remained live for approximately 5 hours and 20 minutes before removal.
The root cause was a flawed GitHub Actions CI workflow contributed via a pull request on August 21, estimated to have been generated by Claude Code. A malicious commit on August 24 modified the CI workflow so that the npm token used for publishing Nx packages would be sent to an attacker-controlled server via webhook.
The malicious payload weaponized local AI coding agents (Claude, Gemini, and q) via a dangerous prompt to inventory sensitive files and exfiltrate secrets, credentials, and sensitive data to a public GitHub repository named s1ngularity-repository with a numeric suffix.
Reported impact
Hundreds of thousands of Nx users unknowingly downloaded the poisoned programs. Attackers directed the malicious code to seek out account passwords, cryptocurrency wallets, and other sensitive data from those who downloaded the altered programs.
Index analysis
Security researchers characterize this as likely one of the first documented cases of malware leveraging AI assistant CLIs for reconnaissance and data exfiltration, combining established supply-chain attack techniques with AI manipulation.
Classification
Sources and evidence
This record was researched and written by the Index, based on the following original reporting.