convez.ai
MCP ServerIntegração & APIsAvançado
jacobsd32-cpu

DJD Agent Score MCP

jacobsd32-cpu/djd-agent-score-mcp

API de reputação para agentes IA com scoring e detecção de fraude

Ver no GitHub
InicialPluginScore · de 0 a 100

MCP server que expõe a API de reputação DJD Agent Score para agentes de IA. Fornece ferramentas nativas para scoring de wallets, verificação de fraude e gestão de reputação em blockchain (Base), com suporte a micropagamentos x402.

Instalação

$npx convez-ai jacobsd32-cpu/djd-agent-score-mcp

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
1
Acompanhando
0
Saúde
42%
Último commit
há 8 meses
Versão
v1.1.0
Issues abertas
1
Contribuidores
2

Casos de uso

  • Avaliar confiabilidade de wallets em agentes autônomos
  • Detectar atividades fraudulentas em transações blockchain
  • Integrar sistema de reputação em plataformas de IA descentralizadas

Veredito

Server robusto e bem documentado para integração de scoring de reputação blockchain em agentes IA. Requer compreensão de x402 micropagamentos e wallets Base; ideal para aplicações descentralizadas e auditoria de agentes.

README

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

djd-agent-score-mcp

MCP server for DJD Agent Score — a reputation scoring API for AI agent wallets on Base.

Listed in the Coinbase x402 Ecosystem npm version

This server exposes the DJD Agent Score REST API as Model Context Protocol tools, so any MCP-compatible agent (Claude, GPT, Gemini, LangChain, etc.) can call scoring endpoints natively.

Tools

ToolEndpointCostDescription
score_basicGET /v1/score/basicFreeBasic score, tier, confidence
score_fullGET /v1/score/full$0.10 (x402)Full dimension breakdown
score_refreshGET /v1/score/refresh$0.25 (x402)Re-score with latest chain data
report_fraudPOST /v1/report$0.02 (x402)Submit fraud report
check_blacklistGET /v1/data/fraud/blacklist$0.05 (x402)Check fraud reports
get_badgeGET /v1/badge/{wallet}.svgFreeEmbeddable SVG badge
get_leaderboardGET /v1/leaderboardFreeTop scored wallets
register_agentPOST /v1/agent/registerFreeRegister wallet with metadata
health_checkGET /healthFreeSystem status

Installation

npm install -g djd-agent-score-mcp

Or run directly with npx:

npx djd-agent-score-mcp

Configuration

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "djd-agent-score": {
      "command": "npx",
      "args": ["-y", "djd-agent-score-mcp"]
    }
  }
}

Cursor

Add to .cursor/mcp.json in your project root:

{
  "mcpServers": {
    "djd-agent-score": {
      "command": "npx",
      "args": ["-y", "djd-agent-score-mcp"]
    }
  }
}

Claude Code

Add to your project's .mcp.json:

{
  "mcpServers": {
    "djd-agent-score": {
      "command": "npx",
      "args": ["-y", "djd-agent-score-mcp"]
    }
  }
}

Generic MCP Client (Streamable HTTP)

Start the server in HTTP mode:

TRANSPORT=http PORT=3000 npx djd-agent-score-mcp

Then connect your MCP client to http://localhost:3000/mcp.

Environment Variables

VariableDefaultDescription
DJD_BASE_URLhttps://djd-agent-score.fly.devAPI base URL (use http://localhost:3001 for local dev)
DJD_TIMEOUT_MS10000Request timeout in milliseconds
TRANSPORTstdioTransport mode: stdio or http
PORT3000HTTP server port (only used when TRANSPORT=http)

Development

git clone <repo-url>
cd djd-agent-score-mcp
npm install
npm run build
npm start

To point at a local API during development:

DJD_BASE_URL=http://localhost:3001 npm start

x402 Payment

Some endpoints require x402 micropayments. When an agent calls a paid tool, the API responds with HTTP 402 and payment instructions. Your agent framework must:

  1. Detect the 402 response
  2. Complete the x402 payment (USDC on Base)
  3. Retry the request with the payment proof

The MCP server surfaces the 402 details in the tool's error response so the agent can handle it.

License

MIT

Ver o original no GitHub

Bom para

Relacionados