News
What just dropped, what it means, and whether you should care.
How Anthropic's Dreaming Lets Claude Agents Fix Their Own Mistakes
Anthropic's Dreaming feature -- announced May 6, 2026 -- lets Claude agents asynchronously review past session transcripts, extract cross-session patterns, and rewrite their own memory stores between
Claude Code v2.1.138: The Session ID Env Var Builders Needed
Claude Code v2.1.138 ships CLAUDE_CODE_SESSION_ID as an environment variable in every Bash subprocess -- the same session_id your hooks already receive. That means hooks and Bash tool calls can finall
OpenAI Symphony: Linear as a Coding Agent Control Plane
OpenAI Symphony is an open-source Apache 2.0 orchestrator released April 27, 2026, that maps every active Linear issue to a dedicated Codex agent workspace. It polls Linear every 30 seconds, creates a
Gaslightus 4.7: What the Claude Opus Developer Backlash Was Actually About
Claude Opus 4.7 launched April 16, 2026 and earned the nickname "Gaslightus 4.7" within 24 hours of release. The complaints were real -- but the worst behavior came from three separate bugs Anthropic
Google's $40B Anthropic Bet: What It Means for Claude Builders
Google committed up to $40 billion to Anthropic in April 2026 -- $10B immediately at a $350B pre-money valuation, with $30B more tied to performance milestones. Combined with Amazon's $25B commitment
Anthropic's Claude Agents Outperformed Human Researchers 4x
Nine Claude Opus 4.6 agents working in parallel for 5 days recovered 97% of an AI alignment performance gap. Two senior human Anthropic researchers recovered 23% over 7 days on the same problem. Total
Self-Evolving Agents Are Here: What Builders Need to Know
Self-evolving AI agents rewrite their own code, prompts, and skills after each run -- without human intervention. Two open-source projects hit viral GitHub traction in April 2026: EvoMap's Evolver (Ge
Mastercard Agent Pay: What It Actually Means for Builders
Mastercard's Agent Pay lets AI agents execute verified purchases on your behalf using a framework called Verifiable Intent -- a cryptographic record linking who authorized the action, what instruction
MCP Explained: What Model Context Protocol Is and Why Every AI Agent Builder Needs to Know It
If you've been building with AI agents for more than a few months, you've probably seen "MCP" mentioned more and more. In Anthropic community calls, in OpenClaw documentation, in developer threads. It started as a technical detail and it's quickly becoming something every serious AI builder needs to understand. MCP stands for Model Context Protocol. And it's about to change how AI agents work with external tools and data in a fundamental way. Let me break it down without the developer-speak.
The AI Agent Landscape: Every Platform You Need to Know About in 2026
The AI agent space has exploded. A year ago there were a handful of serious tools. Now there are dozens of platforms, each claiming to be the one you need. Most lists you'll find are either outdated or written by people who've never actually used the tools. This one is different. I've personally used most of these. For the ones I haven't run hands-on, I've dug into documentation, real user reports, and developer feedback. Here's the full picture as of 2026. What We Mean By "AI Agent Platform
Google Gemini Agents: Everything You Need to Know
Google is betting big on AI agents. Gemini is no longer just a chatbot. With the launch of Gemini agents (previously called "Project Mariner" and "Jules"), Google has entered the autonomous AI agent space. Here is everything you need to know as of April 2026. What Are Google Gemini Agents? Gemini agents are autonomous AI systems built on Google's Gemini models that can browse the web, use Google Workspace tools, execute code, and complete multi-step tasks without constant human input. They re