npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@expertcustom/aurora-mcp

v0.5.1

Published

MCP server to connect Claude Code (and other MCP clients) to Aurora AIs

Downloads

1,208

Readme

@expertcustom/aurora-mcp

Servidor MCP (Model Context Protocol) que conecta Claude Code — e qualquer outro cliente MCP — às IAs do Aurora. Permite listar, inspecionar e conversar com as IAs configuradas no seu workspace Aurora diretamente do seu editor.

Como funciona

Claude Code  ──stdio──>  npx @expertcustom/aurora-mcp  ──HTTPS+JWT──>  Aurora backend

As credenciais do provider (OpenAI/Claude/Gemini) ficam no backend, vinculadas a cada IA. O MCP nunca enxerga essas chaves — ele só repassa sua identidade Aurora (JWT) e o backend faz a chamada com a credencial da IA.

Pré-requisitos

  • Node.js ≥ 20
  • Conta no Aurora com permissão ia:read
  • URL do backend Aurora (ex: https://ia.api.expertcustom.com.br)

Instalação

1) Login no Aurora

Uma vez por máquina:

npx @expertcustom/aurora-mcp login

O comando pergunta:

  • Aurora URL — endpoint do backend (ex: https://ia.api.expertcustom.com.br)
  • Email — seu email cadastrado no Aurora
  • Senha — sua senha

O JWT fica salvo em ~/.config/aurora-mcp/credentials.json (modo 0600). Pra trocar de conta ou refazer o login, é só rodar de novo.

2) Registrar no Claude Code

Forma mais simples:

claude mcp add aurora -- npx -y @expertcustom/aurora-mcp

Ou edite manualmente o ~/.claude.json, seção mcpServers:

{
  "mcpServers": {
    "aurora": {
      "command": "npx",
      "args": ["-y", "@expertcustom/aurora-mcp"]
    }
  }
}

3) Verificar

Dentro do Claude Code, rode /mcp — deve aparecer aurora como connected.

Tools disponíveis (86)

| Categoria | Tools | |-----------|-------| | IAs | list_ais, get_ai_config, create_ai, update_ai, delete_ai | | Chat | chat_with_ai | | Conversas | list_conversations, get_conversation, delete_conversation, get_conversation_status, resume_conversation, close_conversation, handoff_ack, list_recent_conversations | | Skills | list_skills, get_skill, create_skill, update_skill, delete_skill, import_skill, export_skill, approve_skill, reject_skill, move_skill | | Skill Files | list_skill_files, get_skill_file, create_skill_file, update_skill_file, delete_skill_file, upload_skill_file | | Embeddings | list_embeddings, create_embedding, update_embedding, delete_embedding, delete_all_embeddings, get_embedding_stats | | Documents | list_documents, get_document, delete_document, upload_document | | UI Actions | list_ui_actions, get_ui_action, create_ui_action, update_ui_action, delete_ui_action, link_ui_action_to_ai, unlink_ui_action_from_ai, ui_action_stats | | MCP Servers | list_mcp_servers, get_mcp_server, create_mcp_server, update_mcp_server, delete_mcp_server, link_mcp_server_to_ai, unlink_mcp_server_from_ai | | Providers | list_providers, get_provider, create_provider, update_provider, delete_provider, test_provider | | Scheduled Tasks | list_schedules, get_schedule, create_schedule, update_schedule, delete_schedule | | Usuários | list_users, get_user, create_user, update_user, delete_user, transfer_user_resources, get_me | | Roles | list_roles, list_permissions, create_role, update_role_permissions, delete_role | | Settings | get_settings, update_settings | | Audit | list_audit | | Token Usage | token_usage | | Webhook Inspector | read_webhook_inspector, clear_webhook_inspector | | Status | get_status |

Todas as tools operam via API REST do Aurora. Credenciais de provider nunca são expostas.

Outros clientes MCP

Cursor, Continue, Cline e qualquer outro cliente MCP usam o mesmo formato — basta apontar o command para npx e os args para ["-y", "@expertcustom/aurora-mcp"].

Trocar de ambiente / fazer logout

Refaça o login apontando para outra URL:

npx @expertcustom/aurora-mcp login

Pra remover as credenciais:

rm ~/.config/aurora-mcp/credentials.json

Exemplos: Agendar Tarefas com a IA

A tool schedules permite que a IA crie, liste e gerencie tarefas agendadas. Exemplo de conversa:

Usuário: "Agende uma tarefa para gerar o boletim todo dia às 8h"

Claude (usando schedules tool):
  create_schedule({
    aiId: "aurora-123",
    title: "Gerar boletim diário",
    instruction: "Gere o boletim com dados de hoje",
    deliveryType: "email",
    target: "[email protected]",
    preset: { kind: "daily", hour: 8, minute: 0 }
  })

Resultado: Tarefa criada e agendada ✓

Tipos de entrega:

  • email — envia resultado por email (requer target com email válido)
  • whatsapp — envia no WhatsApp (requer target com número/chatId e IA com provider ativo)
  • internal — resultado fica no histórico da conversa (sem target)

Presets de recorrência:

  • { kind: "daily", hour: 8, minute: 30 } — todo dia às 8:30
  • { kind: "weekly", weekday: 3, hour: 14, minute: 0 } — quarta-feira às 14:00
  • { kind: "monthly", day: 1, hour: 9, minute: 0 } — 1º de cada mês às 9:00
  • { kind: "hourly", everyHours: 3, minute: 15 } — a cada 3 horas na marca :15

Mais exemplos no dashboard Aurora → aba "Agendamentos".

Troubleshooting

  • No Aurora credentials found — rode npx @expertcustom/aurora-mcp login primeiro.
  • Aurora API error (401) — JWT expirou ou inválido. Refaça o login.
  • Aurora API error (403) — sua role não tem ia:read. Fale com um admin.
  • Aurora API error (404) em chat_with_ai/get_ai_configname da IA não existe ou não pertence ao seu workspace. Confirme via list_ais.
  • Aurora API error (409) em chat_with_ai — a IA está inativa ou sem credencial de provider. Ajuste no dashboard.