Run any Skill in Manus with one click

Get Started

widget-inference

Stars1

Forks0

UpdatedJuly 6, 2026 at 23:31

Guidelines for mathematically deducing advanced widget data on the backend.

Installation

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

Run Skill in Manus

Source

agussantinelli

agussantinelli/GitData

View GitHub Repository View Creator Repositories

Download

Run Skill in Manus

Related occupationsSOC

Based on SOC occupation classification

Software DevelopersComputer and Mathematical Occupations·SOC 15-1252

SKILL.md

readonly

name	widget-inference
description	Guidelines for mathematically deducing advanced widget data on the backend.

Widget Inference Engine

Context

GitData's premium UI (like Tech Radar, Achievements, and Dev Clock) requires data that GitHub does not natively provide. Instead of adding complex databases, GitData infers these states algorithmically in the backend (OctokitGithubRepository.ts) using the raw metrics available.

Guidelines

Tech Radar Categorization: Languages must be statically mapped into three core pillars: Frontend, Backend, and DevOps/Data. The percentage is calculated by summing the total bytes of languages matching these buckets.
Achievements Deduction: Trophies are awarded based on absolute thresholds of existing data (e.g., prs >= 5 -> Pull Shark, followers >= 10 -> Influencer).
Time Analysis: Time-based widgets (Dev Clock, Hourly Frequency) rely exclusively on the created_at timestamp extracted from the last 50 REST API publicEvents.
Backend Processing: All inference MUST happen in the Backend. The Frontend should only receive the finalized, computed structures (e.g., arrays of objects) and should NOT perform heavy mapping or categorization logic.

Examples

// Inferring an achievement
const achievements = [];
if (prs >= 5) achievements.push({ id: 'pull-shark', title: 'Pull Shark', description: '...', icon: '🦈' });

// Inferring Tech Radar
let frontend = 0;
const feLangs = ['HTML', 'CSS', 'JavaScript', 'TypeScript', 'React'];
Object.keys(languageMap).forEach(lang => {
  if (feLangs.includes(lang)) frontend += languageMap[lang];
});

More from this repository

same repository

node

agussantinelli/GitData

Node.js best practices and runtime configuration for GitData-BackEnd.

2026-07-081

frontend-test-enforcement

agussantinelli/GitData

Mandatory rule requiring a corresponding Vitest/React Testing Library test file for every React component (.tsx) in GitData frontend.

2026-07-071

global-skills

agussantinelli/GitData

Catalog of all specialized skills used across the GitData Monorepo project.

2026-07-061

github-api-hybrid

agussantinelli/GitData

Rules for using both GraphQL and REST APIs simultaneously for GitHub data fetching.

2026-07-061

widget-i18n-architecture

agussantinelli/GitData

Architecture rules for building UI widgets, supporting i18n locales, and multi-theme logic.

2026-07-061

css-structure

agussantinelli/GitData

Mandatory rule to keep CSS files inside a styles/ folder relative to their component.

2026-07-061

name	widget-inference
description	Guidelines for mathematically deducing advanced widget data on the backend.

Widget Inference Engine

Context

Guidelines

Tech Radar Categorization: Languages must be statically mapped into three core pillars: Frontend, Backend, and DevOps/Data. The percentage is calculated by summing the total bytes of languages matching these buckets.
Achievements Deduction: Trophies are awarded based on absolute thresholds of existing data (e.g., prs >= 5 -> Pull Shark, followers >= 10 -> Influencer).
Time Analysis: Time-based widgets (Dev Clock, Hourly Frequency) rely exclusively on the created_at timestamp extracted from the last 50 REST API publicEvents.
Backend Processing: All inference MUST happen in the Backend. The Frontend should only receive the finalized, computed structures (e.g., arrays of objects) and should NOT perform heavy mapping or categorization logic.

Examples

// Inferring an achievement
const achievements = [];
if (prs >= 5) achievements.push({ id: 'pull-shark', title: 'Pull Shark', description: '...', icon: '🦈' });

// Inferring Tech Radar
let frontend = 0;
const feLangs = ['HTML', 'CSS', 'JavaScript', 'TypeScript', 'React'];
Object.keys(languageMap).forEach(lang => {
  if (feLangs.includes(lang)) frontend += languageMap[lang];
});