Hacker News: michelhabib

New comment by michelhabib in "Show HN: Gnosis, a knowledge base for what the code can't tell you"

michelhabib — Mon, 27 Apr 2026 15:07:16 +0000

Im currently using mintlify, with a section for design decisions, and why i picked evey one. usually i start by discussing the design, comparing options, picking a suitable one with a certain criteria, document it all in an md, and add it to that section. what is the workflow in your case?

New comment by michelhabib in "Show HN: OSS Agent I built topped the TerminalBench on Gemini-3-flash-preview"

michelhabib — Mon, 27 Apr 2026 14:57:30 +0000

woow, looks very good. I'm wondering if you do any optimizations for cli in general, since you're not using MCP. I'm building my own CLI for AI Agents, and was always concerned with context rot.

New comment by michelhabib in "Show HN: AI memory with biological decay (52% recall)"

michelhabib — Mon, 27 Apr 2026 14:51:13 +0000

hahaha, that's true. Intelligent and human may not be efficient, as humanity is much flawed by design

New comment by michelhabib in "Show HN: Browser Harness – Gives LLM freedom to complete any browser task"

michelhabib — Sun, 26 Apr 2026 17:09:29 +0000

Absolutely stunning. Do you think Browsers and websites will fight back to slow down agent's upcoming control of the browser? I can think of a 1000 ways to use this. A website can think of 1000 reasons not to let me do so.

New comment by michelhabib in "Show HN: A Karpathy-style LLM wiki your agents maintain (Markdown and Git)"

michelhabib — Sun, 26 Apr 2026 16:59:09 +0000

Just wondering if AI Notes would add value or create noise. I do love the website Ascii style :)