Agents
Latest articles about agents.
AgentsSuspected Kazakhstan-Linked Hackers Used AI Agent to Spy on Central Asian GovernmentsCloudSEK researchers say the Silent Lynx group used an open-source AI agent called Hermes, powered mainly by DeepSeek models, to automate parts of an espionage campaign across Central Asia.AgentsShow HN: Senku, an AI agent that controls your phone in the backgroundSenku is an Android AI agent that operates apps on a hidden screen while the user keeps using the phone. It is distributed as an APK, with iOS support planned and pricing tiers up to $100 per month.
AgentsClosed: an AI agent for mechanical mould designclosed is a desktop AI agent for mechanical mould design that drafts every mould stage, shows the provenance of each dimension and waits for human approval before anything is released.AgentsHow a Team of Agents Automates OpenTelemetry Instrumentation for GenAI FrameworksAn OpenTelemetry blog post describes a multi-agent system that inspects GenAI frameworks, plans and writes instrumentation against the GenAI semantic conventions, and validates the telemetry with Weaver's live-check.AgentsMudGym Brings MUD2 to the Browser via WebAssembly as a Reinforcement Learning EnvironmentA developer has compiled MUD2, one of the earliest online multiplayer games, to WebAssembly and paired it with MudGym, a reinforcement learning environment whose marimo notebook examples run in the browser through Pyodide.
AgentsExpertCrowd-SWE: Expert-Authored Benchmark Puts Seven Frontier Coding Agents 35 Points ApartA Toptal ExpertCrowd technical report introduces EC-SWE, 100 engineer-written software tasks graded like maintainer review; seven frontier models span 35 points, with Claude Opus 5.5 and GPT-6 Astra statistically tied at the top.
AgentsAWS describes a seven-layer composable reference architecture for enterprise agentic AI on Amazon Bedrock AgentCoreAn AWS Machine Learning Blog post presents the Pluggable Agentic AI Framework, a seven-layer reference architecture for taking multi-agent applications from prototype to production, plus a retail customer support reference implementation.
AgentsSage-Taiyi: A Metacognitive Architecture for AI-Driven Scientific DiscoveryShanghai Artificial Intelligence Laboratory introduces Sage-Taiyi, a two-level agent architecture that uses metacognitive monitoring and control to guide scientific search, reporting benchmark leads and five expert-guided validations in chemistry, materials, and engineering.