Go

If you read my last post, you know I built searxng-web-fetch-mcp — a Crystal MCP server that gives AI assistants web search and content fetching. It worked. It was fast. I shipped it in a night. But after a few weeks of daily use, things started to bother me. The extraction was good, but not great. Some articles came back with navigation junk. Others had the main content buried under sidebar text. And the batch fetch, while fast, didn’t give me the control I wanted over what came back and how. ...

The last post was about architecture decisions. This one is about execution. I spent this cycle turning ideas into something runnable and testable. Not polished. Not “AI magic.” Just real foundations. What shipped since the last post 1) REST API vertical slice navi-agent now has a working API backbone with health, task, and agent routes, including sync flow. This gave me a full path from request → service → persistence → response, which is where real design flaws start to show up. ...

It was 3 AM when I had the idea for navi-agent a few months ago. I was in bed thinking about the impact of LLMs on developers’ hard skills. Before the LLM boom, I improved my coding skills by building projects for my own use. But when OpenAI launched GPT-3, I saw that this technology could be useful. I spent a lot of time playing with GPT-3 code generation, and I remember the feeling I had when I used it to learn OOP. I was like, “What the F*! How the f* do these guys do that?” That was the spark that made my hyperfocus kick in to study the area. ...

Hello world guys! The TL;DR is: I’ve tested openclaw and other AI orchestrators, and they always follow the exact same pattern: They are built as products to be sold, not as open-source projects for the community. They are created by the hype and for the hype, pushing a generic idea of “agency”—a bloated product with a bunch of features and skills that, at the end of the day, aren’t even that useful. That’s because they aren’t built to solve real problems; they’re built for marketing and to sell big tech subscriptions. That’s exactly why OpenAI hired Peter Steinberger—a classic acqui-hire just to have another avenue to sell their API keys and subscriptions, not to solve actual problems. ...