@ankrgyl Datadog. They have an MCP and a relativley new CLI (pup). CLI is objectively better because of composition (you really can’t analyze large data effictively without it).
This is a very well implemented CLI. It's so thorough. It dynamically registers commands, it's designed for a browser-wielding agent to automate the setup steps, it can start a MCP daemon…
The author wrote a really cool article:
https://t.co/xqWmekmTA3
@badlogicgames I thought so. But my day to day still involves people pushing for building MCP replacements for CLI’s that already do the job. So I guess not 🤷
The best thing about this next phase of AI is how it's reviving early Unix/Web values: small, sharp, composable CLI tools and plain text. It’s the environment agents thrive in.
Almost any software that does not work well with AI agents, is not going to make it.
I’m using a lot of tools with tons of UI, all of it will either have to be rewritten to be agent friendly, or it’s dead. For a lot of it too, slapping some MCP on it won’t be enough either.
Really dissatisfied with @apple hardware these days. Second Macbook Air in a year where the screen won't turn on. Maybe this is the year of the Linux desktop (for me).
e.g. can't have a workflow that handles github deployments, and another one that creates them (e.g. on push, or on issue label). The deployments created through Actions won't trigger the deployment handler. Shame.
Hey #YangGang! I just stepped off the stage in Detroit, and wow, what a fantastic night. I hope I made you proud, I think the Yang Gang is about to get a lot bigger. If you can, please make a donation tonight, and then head out to celebrate this win! https://t.co/daodQfDFvt