The stealer ecosystem has matured into a professionalized criminal economy that most organizations are simply not monitoring closely enough.
While the industry fixates on household names like Lumma and RedLine, a growing class of lesser-known, actively deployed stealers, Void, a C++ infostealer that emerged in late 2025, Datura, Misericorde, Saturn, and others, are quietly collecting credentials, session cookies, and crypto wallet data from victims worldwide, feeding logs into underground markets that fuel ransomware, account takeovers, and business email compromise.
In a just-released research report The Unknown Stealers: From Dark Web to Log Markets, SOCRadar researchers identify up to six simultaneous active campaigns running on the Void infrastructure. Each campaign used slightly modified binaries, a natural artifact of different affiliates configuring their own builds, but all shared the same underlying C2 relay architecture and Steam-based resolution mechanism. Some Steam accounts used in earlier campaigns had already been deleted, indicating active infrastructure rotation. Void is a textbook example of how low-profile, under monitored stealers can operate at scale before anyone is paying attention.
You can read the research report here: https://socradar.io/resources/whitepapers/stealer-dark-web-log-markets
CData on Claude Managed Agents: Anthropic’s Bet on the Meta-Harness
Posted in Commentary with tags CData on April 13, 2026 by itnerdIn a new blog post, Amit Naik, VP of Artificial Intelligence at CData, explores Anthropic’s “Claude Managed Agents” and what the concept of a “meta-harness” reveals about the next phase of enterprise AI. While much of the market focus remains on model performance, Naik argues that the real shift is happening at the infrastructure layer that enables agents to operate reliably at scale.
The post examines how managed agent platforms abstract the complexity of orchestration, memory, security, and tool integration, allowing organizations to accelerate development without building everything in-house. At the same time, Naik highlights key trade-offs, including potential vendor lock-in and reduced control over data and agent behavior.
In Naik’s opinion, managed agent infrastructure is a critical battleground for enterprise AI, where success will depend not just on model quality, but on how effectively organizations can operationalize and scale intelligent agents.
Read the full blog here: https://www.cdata.com/blog/claude-managed-agents-anthropic-meta-harness
Leave a comment »