AI Agents
AI agents are LLM-powered systems that don't just answer — they plan steps, call tools and drive a task to completion: writing code, searching the web, booking, analyzing. 2026 became the year of agents, from coding agents to autonomous researchers. This page gathers all our coverage of agentic AI: launches, protocols (MCP, A2A), reliability and real-world use.
Claude Sonnet 5 Became First Model to Pass All GitLab Duo Agent Platform Tests
Anthropic released Claude Sonnet 5 on GitLab Duo Agent Platform — the first model to pass all GitLab benchmark tasks, versus 93.8% for Sonnet 4.6.
How Context Engineering Makes Weak Local LLM a Reliable Agent
Author on Habr showed: for local LLM on closed loop, agent quality is determined not by model size but by how its context is structured.
Symlink Vulnerability Affects Six AI Code Agents, Including Amazon Q and Cursor
Wiz found a Unix symlink trick that bypasses the protection of six AI code agents, including Amazon Q and Cursor, and gives attackers access to a developer's…
Meta restricted engineers' use of Claude Code and Codex due to distillation risk
PWN AI expert Artyom Semenov: how AI agents delete files and compromise themselves
AI Agents in Production: Six Architectural Errors That Prevent Them From Reaching Launch
Microsoft Research Unveiled SkillOpt — Learnable Skills for AI Agents
How to Deploy AI Agents in Legacy Code to Avoid Silent Regressions in Production
Habr Translated a Guide on Deploying AI Agents in Legacy Systems: Testing, Documentation, and Compromise Boundaries to Prevent Agents from Reverting Refactoring by…
HashiCorp Released Terraform MCP Server 1.0 — an AI Bridge to Terraform Registry
HashiCorp Moved Terraform MCP Server to GA: the Server Connects LLM Agents with Terraform Registry so AI Writes Current HCL Configurations Rather than Based on…
Sakana AI Released Sakana Marlin — an Agent for 100-Page Reports with Slides
INFERA AI.Firewall and Agent Runtime Security approach for protecting business AI agents
How a Habr author created an AI assistant on Hermes Agent for family and support in Telegram
LangChain and NVIDIA Release NemoClaw Deep Agents Blueprint for Enterprise AI Agents
NVIDIA Developer Blog Explains AI Agent Reinforcement Learning Techniques
NVIDIA Developer Blog published a breakdown of reinforcement learning techniques for AI agents—from classic RLHF to new feedback-based model training approaches.
British Furniture Maker Prepares for Era of AI Bots Buying Sofas
The Cotswold Company faces a new reality: furniture is increasingly being "purchased" not by human buyers, but by AI agents carrying out automatic purchases.