How many AI agents are running inside your org right now?
Ox Security dropped a report on April 15 calling it "the mother of all AI supply chains.
Last month I watched a team debug their customer-support agent for three days. Hallucinations, wrong tool calls, invented parameters.
Pluto Security calls it MCPwn, which is about as on-the-nose as vulnerability names get.
Somewhere right now, a team lead is telling their CTO that swapping LangGraph for CrewAI will take "a sprint, maybe two.
OpenAI shipped a Codex update on April 16 that got buried under the usual cycle of model-release drama and benchmark wars.
I spent last week trying something simple: "Claude, draft me a blog post about Docker network modes and publish it." The drafting part worked.
Twelve months ago, most agent teams cared about one protocol: MCP. It handled the plumbing between an agent and its tools, and that was enough.
Last month, researchers at Token Security dropped a vulnerability report that should have made every MCP server operator lose sleep.
Ninety-seven million installs. That's where MCP stood at the end of March, and by any measure, the protocol has won.
Sixteen months ago, connecting an AI agent to your company's tools meant writing bespoke integrations for every provider. OpenAI had function calling.
Somewhere around Day 16 of building Postlark, we published an npm package and accidentally created our most effective distribution channel.
Most blog posts go live the instant they're done.
A year ago, people were debating whether MCP would become the standard for connecting LLMs to tools. That debate is over — MCP won.
If you've been anywhere near developer Twitter or Hacker News this quarter, you've seen OpenClaw.
Five days ago, Figma flipped a switch that most designers haven't fully processed yet.
Somewhere around Day 15 of building Postlark, I shipped a feature that most people would consider strange for a blog platform. Not themes.
I tried to automate my blog last month. The writing part was easy — Claude handled that in seconds.