GitHub Trending Intelligence Report

Problem

GitHub Trending shows which repositories are gaining attention, but a copied leaderboard does not explain why they are trending, what category they belong to, or whether they are worth inspecting. Manual summarization can also blur verified fields such as stars, language, description, and repository URL.

How It Works

The skill treats GitHub Trending as a structured data source. It infers filters such as language, daily/weekly/monthly, and spoken_language from the request, then runs scripts/fetch_trending.py to obtain JSON. The report uses script fields for facts, including total_stars, current_period_stars, primary_lang, description, and url. The analysis layer classifies projects as AI agents, developer tools, frameworks, infrastructure, security, UI, or related areas, and forms grounded hypotheses for why they are trending.
The output starts with title, date, range, language, and spoken language, then an Overview table, followed by concise per-project notes on capability, technical signal, and trend hypothesis. If the user asks for JSON, it returns script JSON or a strict subset; if no repositories match, it states that explicitly and lists the filters used.

Boundaries

It is useful for trend scanning, direction inference, and candidate selection, but not for code audit, security assessment, or dependency selection. Inferences are limited to repository descriptions, languages, and public context. Missing information is marked not specified or omitted, and speculation should not be presented as fact.