How do you know the agent isn't quietly rotting your codebase?
Agent-written code passes review and still erodes a system. Four mechanisms in Kiwipow Agent that answer it with evidence rather than with trust.
Latest posts
How an agent verifies a webpage it can never look at
mount, flush, fakeServer and mountRouting. The four helpers that let an agent verify a page the way the app uses it, from vitest, without a server or a browser.
Stop counting duplicated lines: I rank copy-paste by what it costs me
An MCP server that finds duplicated code by meaning, ranks it by what it costs to keep, and hands the worst of it to my AI agent.
A blank page and a green test: the bug an agent can't see
A template that cannot resolve a path renders an empty string. For a human that is a blank spot on the page. For an agent it is a green test. Here is how the…
No virtual DOM, nothing re-renders: the whole model in four rules
Plain custom elements, a synchronous lifecycle, nothing re-renders on its own, and typed events. Four rules, each chosen because a reviewer can check it by…
Skills, not docs
Documentation tells an agent how a library works. A skill tells it what it will get wrong from habit. Keeping the two apart is what makes either one useful.
Why I stopped handing agents a framework
Coding agents write better Vue and Angular than they will ever write for my library. I use the library anyway, and this is the honest version of why.
Live API specs for coding agents
Live API specs for coding agents An agent writing frontend code has to know the backend's...
Giving AI agents knowledge they were never trained on
I love coding my own stuff, and my clients typically have lots of internal specifications and...
Struggling with Visual Studio 2017, xUnit and .NET standard?
Have you created a class library for tests with XUnit and .NET standard? Doesn’t it work?