Follow me and let’s build some AI agents
Daytime quality on Claude. After-hours spend on Grok. Same URL, same "model": "claude". The switch is a CEL virtual-model policy in standalone agentgateway — not an if/else in your app.
Grok Build is xAI's coding agent. Terminal, repo, shell. Out of the box it talks straight to api.x.ai with a key in ~/.grok.
Two days ago DeepSeek open-sourced DeepSeek Harness. The pitch is one line: everything is a plugin. Models, tools, skills, the agent loop, even the UI.
Claude Desktop talks to Anthropic by default. As of 9 July 2026 you can point the whole app (Chat, Cowork, and Code) at a gateway you run yourself.
OpenMausBot is a chat app for a little team of bots. It is not another OpenAI wrapper.
Here’s a familiar pattern. You stand up a few agents, each in its own pod. You give one a supervisor prompt that vaguely says “delegate when needed.” Tools get wired in application code. The dem…
Nobody Signed Off on the AI Bill
agentgateway v1.4.0 landed on July 27, 2026.
You wouldn’t put a hundred microservices into production with no gateway, no auth, no rate limits, and no logs. Yet that is exactly how most teams ship AI agents today: an API key baked into a conta…
OpenClaw’s memory_search and memory_get tools enable semantic recall from MEMORY.md, memory/*.md, and session transcripts before answering about prior work/decisions. Defaults use cloud embeddings, …
We're past the era of chatbots. Today's AI agents don't just answer questions—they do things. They manage calendars, write code, handle emails, and make decisions on your behalf.
Reading Time: 15 minutes Tags: #AI #FortiGate #Kubernetes #MCP #DevOps #NetworkSecurity Published: January 2026