convez.ai
MCP ServerIntegração & APIsIntermediário
agent-tool-intel

Agent Tool Intelligence

agent-tool-intel/agent-tool-intel

Qualidade e descoberta para 39K+ servidores MCP — com transparência

Ver no GitHub
InicialPluginScore · de 0 a 100

Plataforma de avaliação e descoberta de servidores MCP com análise profunda de qualidade, confiabilidade e adoção comunitária. Oferece busca semântica, scoring transparente baseado em 13 dimensões de sinal e dashboard de saúde para construtores de ferramentas de IA.

Instalação

$npx convez-ai agent-tool-intel/agent-tool-intel

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
1
Forks
0
Acompanhando
0
Saúde
57%
Último commit
há 5 meses
Versão
—
Issues abertas
0
Contribuidores
2

Casos de uso

  • Validar qualidade de um MCP server antes de integração
  • Descobrir ferramentas mais confiáveis para agentes IA via busca semântica
  • Monitorar saúde e ranking do próprio servidor MCP
  • Analisar tendências do ecossistema de ferramentas de IA

Veredito

Ferramenta essencial para quem constrói ou integra MCP servers — transforma descoberta de ferramentas de IA de processo cego em decisão baseada em dados transparentes e confiáveis. Ecossistema maduro com API pronta e roadmap bem definido.

README

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

Agent Tool Intelligence

You built an MCP server. Is it any good?

54% of MCP servers have solid code but zero community adoption — invisible to the agents that need them. Don't let yours be one.

39,752+ servers scored. Quality Score + Community Bonus + Trust Bonus. Free. No signup. Open source.


🗺️ You just found treasure. Now what?

You are a...Start here
🛠️ MCP Builder→ Check your server's grade instantly — paste your GitHub URL
🤖 AI Developer→ API Quick Start — one curl to search 39K+ tools
🔍 Just curious→ See the public leaderboard — top tools ranked
🏢 Enterprise / Team→ Scoring Methodology — how we score, transparently
🤝 Want to partner?→ Open an issue or contact us

💎 The Treasure: What's Inside

For MCP Builders

TreasureWhat you get
🏷️ Grade BadgeDynamic SVG badge for your README. Auto-updates. Free forever.
📊 Health DashboardPer-repo score breakdown. See exactly what to improve.
💡 Improvement TipsSpecific, actionable advice to raise your grade.
📈 Monthly Ecosystem ReportWhere your tool ranks. Trends. Hidden gems.
⚡ Execution Analytics (coming)How many agents use your tool? Success rate?

For AI Agents

TreasureWhat you get
🔍 Semantic Search API"Extract tables from PDF" → ranked results. Not keyword match.
🧠 Additive ScoringQuality Score + Community Bonus + Trust Bonus. Clear path to improve.
🛡️ Trust EngineReal-world success rates. Not simulated.
🚀 Execution Gateway (coming)Execute tools via one API call.

For Everyone

TreasureWhat you get
📖 Open SourceMIT license. All scoring methodology public.
🔬 Transparent Scores13 signal dimensions. No black box.
📊 Ecosystem Data39,752 servers. Grade distribution. Trends.

🗺️ Dig Deeper: The Full Map

🥇 Level 1:  Self-check your tool         → https://agent-tool-intel-production.up.railway.app
🥈 Level 2:  Read the API docs            → https://agent-tool-intel-production.up.railway.app/docs
🥉 Level 3:  Understand the methodology   → https://agent-tool-intel-production.up.railway.app/scoring/methodology
🏆 Level 4:  See the roadmap              → https://agent-tool-intel-production.up.railway.app/roadmap
👑 Level 5:  Become a partner             → Open an issue or email us

🚀 Quick Start

# Search for tools (agents call this)
curl -X POST https://agent-tool-intel-production.up.railway.app/api/v1/search \
  -H "Content-Type: application/json" \
  -d '{"query": "extract tables from PDF", "maxResults": 3}'

# Embed a grade badge (builders do this)
# [![Grade](https://agent-tool-intel-production.up.railway.app/badge/YOUR_ORG%2FYOUR_REPO)]
# (https://agent-tool-intel-production.up.railway.app)

📊 By The Numbers

StatValue
MCP Servers Indexed39,752
Tools Scored39,762
Grade A (truly exceptional)~4%
SourcesGitHub, npm, GitLab, Official Registry
Open SourceMIT License

🏗️ Architecture

This is part of the Agent Tool Platform:

ModuleWhat it doesStatus
Agent Tool Intel ← you are hereQuality scoring + search + trust✅ Live
AgentPilotAgent task orchestration + tool registry✅ Live
AutoMineAutomated tool discovery from content🔧

All three share a unified Canonical ID system: tool:{source}:{namespace}/{name}@version


🤝 How to Contribute


📜 License

MIT — use it, fork it, build on it. Just don't spam people with it.


Ver o original no GitHub

Bom para

Relacionados