maxnew

há 2 mesesdiscuss

Absolutely agree. A lot of LLM-driven work is just inflated busywork with little real output. High token usage doesn’t equal genuine productivity, just unnecessary repetitive verification and paperwork.

maxnew

há 4 mesesdiscuss

This seems quite interesting... I'll give it a try

maxnew

há 4 mesesdiscuss

  I've always found it a bit overkill to open heavy software just to generate a quick normal map. Most online tools either have clunky UIs or upload your files to their backend.

I built this using Three.js and client-side canvas processing (Sobel/Scharr filters). Everything happens locally on your machine, so it's fast and private. Features: - Generate Normal, Displacement, AO, and Specular maps. - Real-time 3D PBR preview with shape selection (Plane, Cube, Sphere). - Fine-grained control over strength, blur, and axis inversion. Would love to get some technical feedback on the processing performance!

maxnew

há 6 mesesdiscuss

Hi HN,

I'm the maker of MCPxel (https://mcpxel.com).

The Problem: As an active user of Claude/MCP, I found the ecosystem fragmented. Searching GitHub for "agent skills" returns thousands of results, but many are broken, outdated, or just simple wrappers. Stars often don't correlate with reliability.

The Solution: I built MCPxel to curate and rate these skills.

How it works:

LLM-as-a-Judge: We use an automated pipeline to evaluate skills on 5 dimensions (Clarity, Utility, Quality, Maintainability, Novelty). S-Tier Filtering: We assign grades (S/A/B/C). S-Tier skills are verified to work out of the box. Role-Based Search: Instead of keyword guessing, you can find tools tailored for specific workflows (e.g., "frontend dev"). Tech Stack: Built with Next.js. We used TRAE (AI IDE) and DeepSeek-V3 heavily during development to accelerate the process.

Try it out: Check out the LLM-judged ratings (S/A/B/C). I'd love to know if the scores match your experience with these tools!

maxnew

há 7 mesesdiscuss

Hey everyone,

I remember struggling with Calculus back in the day, specifically with knowing which rule to apply and when.

I decided to build a tool to help with that. It's a Derivative Calculator that focuses on explaining the solution, not just giving the result.

Why I made it: Most calculators are either paid or just dump the final LaTeX string. I wanted something that visualizes the graph and walks you through the Chain Rule, Product Rule, etc., just like a tutor would.

How we built it We built this project using a modern, performance-focused stack: - Framework: Next.js 15 (App Router) for a blazing-fast, server-rendered experience. - Math Engine: We integrated nerdamer for symbolic computation to handle the heavy lifting of differentiation logic. - Visualization: We used function-plot to build a custom interactive graphing engine that renders functions and their derivatives in real-time. - UI/UX: TailwindCSS and Shadcn/UI ensured the interface remained clean, accessible, and distraction-free.

It's completely free and doesn't require any login.

I'm looking for feedback on the "Steps" generation—does the logic follow how you'd solve it manually?

Thanks for checking it out!

Submissions

Show HN: A high-performance, privacy-focused normal map generator in the browser

Show HN: MCPxel – Navigation and rating station for Agent Skills (LLM-judged)

Show HN: A free derivative calculator with step-by-step explanations

comments