#LLM security research
1161 related articles
·4 min
Building AgentOS with Claude Agent SDK: Automating 95% of Development and Operations Work
Developer Danny Postma built AgentOS on Claude Agent SDK, automating 95% of coding and ops tasks. Deep dive into container isolation, permission control, task orchestration, and human-in-the-loop design.
Read more →
·3 min
Unsloth Desktop Complete Review: An All-in-One Solution for Local AI Model Deployment, Inference, and Fine-Tuning
In-depth review of Unsloth Desktop covering local LLM deployment, inference acceleration, model fine-tuning, multimodal generation, and Agent integration with Claude Code and Codex.
Read more →
·5 min
AI Penetration Testing Learning Roadmap: A Four-Stage Progressive Guide from Beginner to Enterprise-Level Practice
A systematic breakdown of the four-stage AI + penetration testing learning roadmap, covering Agent fundamentals, Web vulnerability discovery, enterprise automation, and advanced practice.
Read more →
·3 min
The Risk of AI Agents Going Rogue: When Autonomous AI Enters Critical Infrastructure
AI agent LeChaton was found in the wild raising safety concerns. This article analyzes threats AI agents pose to critical infrastructure, exploring alignment issues, autonomy risks, and layered defense strategies.
Read more →
·3 min
AI Agent Security Regression Testing: Automatically Blocking Behavioral Degradation Like CI Failures
Traditional quality scores can't catch silent degradation in AI Agent tool-calling paths. Learn how to integrate Agent security regression into CI pipelines by freezing Prompts, model configs, tool schemas, and execution traces.
Read more →
·4 min
Top 10 Hottest AI Skills in the Programming World: From Methodologies to Workflows Fully Explained
In-depth review of the 10 hottest AI Skills in programming, covering Super Powers, MCP Builder, Composio and more—how Skills enable structured workflows, automated testing, and external connections.
Read more →
·3 min
NullOrigin Open-Source Tool: A Deep Dive into the Local Proxy That Removes AI Watermarks and Provenance Metadata
Deep dive into NullOrigin, an open-source local proxy that destroys text watermarks via local SLM rewriting, strips C2PA/EXIF image metadata, and scans for Trojan Source code vulnerabilities.
Read more →
·3 min
LangChain in Practice: A Complete Guide to Building Safe and Controllable AI Agents
Deep dive into LangChain's three-layer ecosystem (LangGraph, LangChain, DeepAgents), with detailed coverage of Guardrail and HITL safety strategies for building secure, controllable AI Agents.
Read more →
·3 min
Thomson Reuters Builds Its Own Frontier LLM: Why Vertical AI Is Becoming the New Battleground
Thomson Reuters launches a proprietary frontier LLM, leveraging decades of legal, tax, and compliance data to build a vertical AI moat. Analysis of the strategy, value, and challenges.
Read more →
·3 min
JetBrains Empowers Local Qwen Deployment on Mac: A Comprehensive Developer Experience Upgrade
JetBrains tooling makes local Qwen LLM deployment on Mac simpler. Explore privacy benefits, cost analysis, and engineering practices for running open-source models on Apple Silicon.
Read more →
·3 min
Inference Engine Vulnerabilities: How LLMs Can Reverse-Control Host Machines — Attack Principles and Defense Strategies
Analysis of LLM inference engine security vulnerabilities, exploring how model outputs can trigger buffer overflows to reverse-control host machines, with defense strategies including sandboxing and Rust.
Read more →
·3 min
Claude Code Enables Auto Mode by Default: A Deep Dive into AI Coding Agent Permission Changes
Anthropic announces Claude Code will default to auto mode from August 14. Analyze the impact on developer workflows, community safety debates, and how to configure permission boundaries.
Read more →
·3 min
Gitar: An AI Code Review Tool That Doesn't Just Find Issues — It Fixes Them
Gitar is an AI code review tool that automatically fixes issues it finds, supports PR review & repair, CI failure diagnosis, and Flaky Test handling. Now part of Sonar.
Read more →
·3 min
Contrive: A Deep Dive into the Unified Cross-App Search and Instant Action Tool for Enterprises
Contrive is an enterprise AI search tool that enables unified search and instant actions across Slack, Notion, Google Drive and more via a keyboard-first command bar.
Read more →
·4 min
Software Engineering in the Agentic Era: The Development Paradigm Is Being Fundamentally Reshaped
An in-depth analysis of how AI agents are reshaping software engineering paradigms—from code completion to autonomous execution—covering agentic workflows, productivity shifts, reliability challenges, and the evolving role of engineers.
Read more →
·4 min
Harness Engineering: A New Paradigm for Multi-Agent Architecture Explained
Deep dive into Harness Engineering's seven core capabilities including tool calling, memory, planning, execution loops, and sandbox security. Learn the evolution from Prompt Engineering to Context Engineering to Harness Engineering.
Read more →
·3 min
How to Choose an AI Agent Framework? Azure AI, Semantic Kernel, and AutoGen Compared in Practice
A systematic comparison of Azure AI Agent Service, Semantic Kernel, and AutoGen—three major AI Agent frameworks—covering positioning, use cases, and hands-on code examples to help developers make the right choice.
Read more →
·3 min
Amazon Bedrock Converse API in Practice: A Development Guide to Unified Interfaces and Streaming Chat
A deep dive into Amazon Bedrock's Converse API unified multi-model interface and ConverseStream streaming output, covering message structures, multi-turn conversations, event stream handling, and AWS ecosystem integration.
Read more →
·3 min
A Single Denominator Caused a Data Leak: The Hidden Trap in a Quantitative Finance Paper
A real data leakage case in quantitative finance: a feature denominator using full-day volume leaked future information into intraday data. Learn the difference between generation and selection leakage.
Read more →
·4 min
LangChain MCP Integration in Practice: From Concepts to Building Multi-Tool AI Applications
Complete guide to integrating LangChain with MCP protocol, covering Agent principles, MCP Server/Client communication, tool reuse, and framework decoupling for building multi-tool AI applications.
Read more →