AI Pulse · Under the Radar

Microsoft just released a tool called skill-recorder on GitHub, and almost nobody noticed. It has 692 stars in four days, which is fast for a developer tool, but there’s been no press coverage. The repo sits under the official Microsoft organization, which makes the silence strange.

Here’s what it does. You record yourself doing a task on screen once. The tool watches, then reconstructs that session into a reusable skill you can plug into Copilot Studio. No manual scripting, no flow diagrams. You show it, it builds the automation.

This matters because it’s the same pattern we’ve been building toward with EDNA’s LP skill capture. Record once, reuse everywhere. Microsoft shipping a first-party version validates the approach, but it also means the window for competitive advantage is narrowing. If you’re not already capturing your team’s workflows and turning them into automations, you’re falling behind companies that are.

The lack of press suggests this is either a quiet internal project that escaped into the open or something Microsoft isn’t ready to promote yet. Either way, the star velocity tells you developers are paying attention. When a tool like this gets traction without marketing, it usually means it solves a real problem people have right now.

What to do with this

If you’ve been waiting to see whether workflow automation is worth the effort, this is your signal. The tooling is here, it’s getting easier, and the big players are moving fast. The original report shows how quickly something like this can go from zero to adoption. This is exactly the kind of capability we build into an AI command centre, where capturing and reusing skills becomes part of how your team works, not a side project.