ytrofr

AI Intelligence Hub — AI skill for Claude Code

AI community

AI Intelligence Hub - Track 14 AI sources (GitHub, HuggingFace, MCP, Claude Code, Anthropic Skills & Cookbooks) with full-text search and keyword scoring.

How to install AI Intelligence Hub

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ytrofr/ai-intelligence-hub and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What AI Intelligence Hub does

AI Intelligence Hub - Track 14 AI sources (GitHub, HuggingFace, MCP, Claude Code, Anthropic Skills & Cookbooks) with full-text search and keyword scoring. Port 4444.

Alternatives in AI

Geo SEO Claude — GEO-first SEO skill for Claude Code 3.6k ★
Codeseek — Rust-powered code intelligence CLI for AI coding agents 764 ★
Academic Commercialization Agent — Turn any research paper into a commercialization report — 6 AI agents, TRL/MRL scoring, patent landscape, mark 581 ★

README

AI Intelligence Hub

**Track AI trends from 24 live sources in one dashboard, and turn them into per-project adoption decisions.** GitHub Trending, HuggingFace, MCP Servers, Claude Code Releases, Anthropic Skills & Cookbooks, arXiv, Google AI Blog, Simon Willison, and more — with full-text search and keyword scoring.

[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](https://opensource.org/licenses/MIT) [![Node.js](https://img.shields.io/badge/Node.js-18%2B-green.svg)](https://nodejs.org/) [![Port](https://img.shields.io/badge/Port-4444-blue.svg)]() [![Sources](https://img.shields.io/badge/Sources-24-purple.svg)]()

Quick Start

git clone https://github.com/ytrofr/ai-intelligence-hub.git
cd ai-intelligence-hub
npm install
npm --prefix web install
npm run build                                          # builds the front end into dist/
cp config/projects.example.json config/projects.json   # then edit with your own projects
node server.js
# Open http://localhost:4444

The front end is a **build**, not a directory of pages. `dist/` is gitignored, so a fresh clone has no UI until `npm run build` produces one — the server says so on startup rather than serving 404s that read like a broken app. While working on it, `npm --prefix web run watch` rebuilds on save and `npm --prefix web run test:watch` is the fast component loop. There is deliberately no dev server: the app is served from the same origin as the API, so there is no proxy, no CORS and no second port.

`config/projects.json` is gitignored — it holds your personal project portfolio. The repo ships `config/projects.example.json` as a template; copy it and customize the `projects` array. The `claude-ecosystem` entry works as-is. `config/radar/*.json` is gitignored for the same reason — see `config/radar/example.json` for the schema.

No paid services required. Most sources are free feeds/APIs; GitHub discovery wants a `GITHUB_TOKEN`. Every fetch reports per-so