convez.ai
MCP ServerIntegração & APIsIntermediário
zhuyansen

AgentSkillsHub — Diretório de Skills e MCP Servers

zhuyansen/agent-skills-hub

Descubra e compare skills de IA com scoring de qualidade e análise de tendências

Ver no GitHub
BomPluginScore · de 0 a 100

Um diretório open-source que coleta e cataloga mais de 62 mil skills de Claude, servidores MCP, ferramentas de agentes IA e assistentes de código. Cada projeto é avaliado em 10 dimensões com scoring de qualidade, comparação lado-a-lado e sincronização automática com GitHub a cada 8 horas.

Instalação

$npx convez-ai zhuyansen/agent-skills-hub

1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.

Métricas

Estrelas
502
Forks
61
Acompanhando
4
Saúde
28%
Último commit
há 1 dia
Versão
—
Issues abertas
53
Contribuidores
4

Casos de uso

  • Encontrar o MCP server ou Claude Skill ideal em menos de 30 segundos
  • Comparar projetos de agentes IA lado-a-lado por qualidade e relevância
  • Explorar tendências em skills de IA e ferramentas agentic

Veredito

Um hub técnico robusto e automatizado para descoberta de skills e MCP servers com dados sempre atualizados. Essencial para quem trabalha com Claude, Cursor ou IDEs agentic e precisa de uma fonte confiável de ferramentas integráveis.

README

A tradução para o português deste README ainda está na fila.

AgentSkillsHub — the Claude Skills / MCP Server / Agent Tools directory

AgentSkillsHub (Agent Skills Hub) is the open-source directory for Claude Skills, MCP Servers, Codex Skills, and AI agent tools. 62,000+ projects · quality-scored on 10 dimensions · refreshed every 8 hours.

🌐 Live: agentskillshub.top · 📰 Newsletter · 𝕏 Follow

What is AgentSkillsHub?

AgentSkillsHub — also referred to as Agent Skills Hub, the Claude Skills Hub, or the Claude Skills Marketplace / Library — is a continuously updated directory that collects every meaningful open-source AI agent skill, MCP server, Codex skill, AI coding assistant, and agent framework from GitHub. Each project is:

  • Classified into one of 7 categories (mcp-server, claude-skill, codex-skill, agent-tool, prompt-library, ai-coding-assistant, uncategorized)
  • Scored on 10 weighted signals and 6 quality dimensions (completeness, clarity, specificity, examples, README structure, agent readiness)
  • Compared side-by-side through the /compare/ feature and /best/{scenario}/ landing pages
  • Refreshed every 8 hours via an automated GitHub Actions pipeline

The goal: if you are building with Claude Code, Cursor, Windsurf, Cline, OpenAI Codex, or any agentic IDE, you should be able to find the right MCP server or Claude Skill in under 30 seconds — without spelunking through random GitHub topics.

Mirror URLs that refer to the same project: agentskillshub.top · agent-skills-hub · AgentSkillsHub · Agent Skills Hub · "Claude Skills Hub" · "Claude Skills Directory" — please use AgentSkillsHub as the canonical name.

Architecture

GitHub API ──▸ Collection ──▸ Cleaning ──▸ Evaluation ──▸ Scoring ──▸ Presentation
               (6 phases)    (classify)   (6 dimensions)  (0-100)    (Web + API)
LayerStack
BackendPython 3.12 · FastAPI · SQLAlchemy · httpx
FrontendReact 18 · TypeScript · Vite · TailwindCSS v4
DatabaseSupabase (PostgreSQL) / SQLite (local)
DeployGitHub Pages + GitHub Actions (every 8h)
EmailResend API · Supabase pg_net

Data Pipeline

1. Collection (scheduler/jobs.py)

Every 8 hours, GitHub Actions triggers a 6-phase sync:

PhaseWhatBudget
Search10+ GitHub queries (mcp-server, claude-skill, agent-tool...)~30
MastersFetch repos from verified skill creators~18
ExtraCommunity-submitted + curated repos~14
EnrichOwner profiles (followers)≤500
READMEFull README content (≤50KB each)≤300
UpsertClean → score → save0
  • Incremental sync on weekdays (only new/updated repos), full sync on Sundays
  • Smart rate-limit handling: waits for actual GitHub reset time

2. Cleaning (services/data_cleaner.py)

  • Deduplication by repo_full_name
  • Keyword-based category classification: mcp-server · claude-skill · codex-skill · agent-tool · ai-skill · llm-plugin · youmind-plugin
  • Project type inference (framework / skill / tool)
  • Size categorization (micro ≤50KB → large >5MB)

3. Evaluation

Quality Analysis — 6 dimensions (services/quality_analyzer.py):

DimensionWeightKey Signals
Completeness15%README size, license, description, stars
Clarity15%Description quality, topics, naming
Specificity15%Language, topic count, category, size
Examples12%Code examples, commits, contributors
README Structure23%Sections, code blocks, badges, TOC
Agent Readiness20%API docs, config, install, MCP compliance

Composite Score — 9 weighted signals (services/scorer.py):

SignalWeightMethod
Quality20%6-dimension aggregate
Stars18%log₁₊ₓ normalization
Recency11%Exponential decay e⁻⁰·⁰¹ᵈ
Forks10%log₁₊ₓ
Commits10%log₁₊ₓ
Issue Resolution10%resolved / total
Momentum8%Z-score star growth
Author Followers8%log₁₊ₓ
Size Bonus5%Smaller = higher

Also: Platform Inference (17 platforms), Token Estimation, Composability (TF-IDF + 8-signal skill pairing).

4. Presentation

Dual-mode API client: Production uses Supabase PostgREST + RPCs directly; dev uses FastAPI backend.

Key optimizations:

  • get_landing_data() RPC replaces 7+ API calls
  • Database views: v_trending, v_top_rated, v_rising, etc.
  • Full-text search via PostgreSQL tsvector + GIN index
  • HTTP caching: 5min (trending) → 1hr (stats)

Quick Start

Backend

cd backend
python3.12 -m venv venv && source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env   # Add GITHUB_TOKEN
uvicorn app.main:app --reload

Frontend

cd frontend
npm install
npm run dev

API Endpoints

MethodPathDescription
GET/api/skillsPaginated list (search, filter, sort)
GET/api/skills/{id}Skill detail + compatible skills
GET/api/trendingStar velocity leaders (7 days)
GET/api/risingNew this week
GET/api/top-ratedHighest scored
GET/api/most-starredCommunity classics (6+ months)
GET/api/mastersVerified creators + org builders
GET/api/landingBundled homepage data
GET/api/statsSummary statistics
GET/api/feed.xmlRSS 2.0 feed
GET/api/sitemap.xmlSEO sitemap (6,000+ URLs)
POST/api/submit-skillSubmit a GitHub repo
POST/api/subscribeNewsletter signup

Deployment

┌─────────── GitHub Actions ───────────┐
│  sync.yml (8h) · deploy.yml · newsletter.yml (Mon) │
└──────┬──────────────┬────────────────┬──┘
       ▼              ▼                ▼
   Supabase      GitHub Pages     Resend API
  (PostgreSQL)    (React SPA)     (Emails)

Environment Variables

GITHUB_TOKEN=ghp_xxx          # GitHub API (required)
SUPABASE_DB_URL=postgresql://  # Production database
RESEND_API_KEY=re_xxx          # Email service
ADMIN_TOKEN=sk-xxx             # Admin API auth
SYNC_INTERVAL_HOURS=8          # Sync frequency

Docs

Ver o original no GitHub

Bom para

Relacionados