Why I Moved My Agent Knowledge Base Off Local Markdown

<h1> Why I’m Building More Local-First Software </h1> <p>Modern applications often follow the same pattern:</p> <p><strong>Open the app → connect to the internet → send data to a

<p>Most AI agents have amnesia.</p> <p>You close the chat window. They forget everything. Every session starts from zero. Every context window fills up and dumps the earliest

<ul> <li>GitHub repository: <a href="https://github.com/gezilinll/agent-grove" rel="noopener noreferrer">https://github.com/gezilinll/agent-grove</a> </li> <li>Online

<p>Hey, me again. Posted MerMark Editor here a while back — first<br /> post got 120+ GitHub stars and the binaries crossed 7 000<br /> downloads, so thanks to everyone who tried

<p><strong>TL;DR:</strong> I built the same browser agent twice — once with 500 lines of Python, once with 7 lines of JSON. The second one took 5 minutes. The agent harness layer

<p>Markdown has been the undisputed king of developer communication for two decades. It’s the language of our READMEs, our static site generators (like Astro), and our technical

<p><strong>Most mind-mapping tools treat Markdown as an export option</strong> — something you get out at the end, one-way. I wanted the opposite: I wanted my mind map to just be

<p>"Notion vs Obsidian" is the wrong framing — that's app vs app. The real comparison is app vs stack: can a markdown setup give you what Notion gives you, out of files you