 Planulix, Claude Code, and the New Era of AI Agents: What's Happening in the Market in May 2026 The market for AI agents in development is overheated to the limit. Several notable materials have been released just in the past week. **Planulix** — a control center for Claude Code, Cursor, Codex, and Kimi. The author has assembled a unified dashboard where you can switch between AI agents and see what each one is doing. The idea is simple: instead of running between terminals — one remote control. It works via an API proxy and a custom task scheduler. What's important: this is not just another "AI tool." It's an attempt to solve the fragmentation problem — when you have 4 AI agents, each with its own context, and they don't see each other's work. Planulix turns them into an orchestra. **Kimi Code CLI** — the Chinese response to Claude Code from Moonshot AI (company valuation $4.3 billion). Open-source, reads and writes code, executes shell commands. On Habr, there are already discussions that Claude Code can be made 8 times cheaper by connecting Kimi K2.6 instead of Claude. **Ollama + Open WebUI on VPS without GPU** — a practical guide on how to run an LLM locally on a CPU. For those who don't want to pay for cloud APIs. **Our conclusion:** the market is moving towards a multi-agent architecture. One AI agent is yesterday's news. Orchestrating multiple agents, their interaction, and centralized management — that's where the main value is being created now. This is the direction we are working on at ASI Biont. → https://asibiont.com/