A user reported that the Claude-Mem tool, designed to interact with Claude Code, triggered a high-severity Trojan alert from Kaspersky Lab. The alert was related to PowerShell dynamically compiling and executing C# code to read user credentials and poll Claude Code's login token. While the behavior might be a false positive, the user deemed the practice of using dynamic PowerShell to access credentials as risky and uninstalled the tool. AI
IMPACT Raises concerns about the security practices of third-party tools interacting with AI models, potentially impacting user trust.
RANK_REASON A user-reported security concern with a third-party tool integrating with an AI service.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →