Malicious code in ai-texts (npm)

The 'ai-texts' npm package (version 1.0.1) advertises itself as a text helper for AI pipelines but includes a malicious dropper in its source code. Upon module import, it downloads a platform-specific installer script from https://anymeetvia.com/download/{win|mac-x86_64|linux}/install1[.ps1] into the OS temporary directory, sets executable permissions on macOS/Linux, and executes it via PowerShell or bash. The dropper then calls process.exit, preventing further code execution. There is no version pinning or integrity verification of the downloaded installer, allowing the attacker full control over the payload. The malicious installer content is not included in the package and can be changed at any time by whoever controls the domain. The README and exported API do not mention this behavior.

This malicious package can lead to arbitrary code execution on the host system with the privileges of the user running the Node.js process. Because the dropper runs immediately on import and downloads an external payload without verification, it enables remote code execution and potential system compromise. The attacker-controlled domain can change the payload at any time, increasing the risk and unpredictability of the impact.

No official patch or remediation is currently documented. Users should immediately stop using 'ai-texts' version 1.0.1 and remove it from their projects. Avoid importing or requiring this package. Verify dependencies carefully and use trusted sources. Monitor for updates or advisories from the package maintainers or npm registry regarding this issue. Consider using package integrity verification tools and lockfiles to prevent untrusted code execution.