convez.ai
MCP ServerDevOps & InfraestruturaAvançado
linuxus

Vault AI

linuxus/vault-ai

Interface IA para Vault: secrets, auth e políticas via chat natural

Ver no GitHub
InicialPluginScore · de 0 a 100

Interface web React para HashiCorp Vault com integração MCP (Model Context Protocol) para gerenciamento de secrets, PKI, autenticação e políticas assistido por IA. Combina operações tradicionais via API com workflows em linguagem natural através de um proxy MCP conectado à Anthropic.

Instalação

$npx convez-ai linuxus/vault-ai

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
0
Forks
0
Acompanhando
0
Saúde
42%
Último commit
há 5 meses
Versão
—
Issues abertas
0
Contribuidores
2

Casos de uso

  • Gerenciar secrets e credenciais com assistência de IA em linguagem natural
  • Operações de PKI e autenticação com interface web moderna
  • Automação de workflows de políticas e controle de acesso baseado em papéis
  • Ambiente demo com usuários e políticas pré-configurados para testes

Veredito

Solução madura e bem arquitetada que reduz fricção no gerenciamento de Vault através de IA conversacional. Recomendado para equipes DevOps que já usam Vault e buscam acelerar operações de secrets e políticas com interface moderna.

README

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

Vault AI

A React web interface for HashiCorp Vault with hybrid MCP (Model Context Protocol) integration — combining traditional API operations with AI-assisted natural language workflows for secrets management, authentication, and policy operations.

Key Features

  • Secrets Management — Browse, create, update, and delete KV secrets with version history
  • Authentication — Token-based login with session management and automatic renewal
  • AI Chat Assistant — Natural language interface for Vault operations via MCP proxy
  • Role-Based Access — Demo environment with pre-configured users and policies

Architecture

┌─────────────────────┐     ┌─────────────────────┐     ┌─────────────────────┐
│                     │     │                     │     │                     │
│   Vault AI Web      │────►│   MCP Proxy         │────►│   HashiCorp Vault   │
│   (React SPA)       │     │   (Node.js)         │     │   Server            │
│                     │────►│                     │     │                     │
└─────────────────────┘     └─────────────────────┘     └─────────────────────┘
        │                                                        ▲
        └────────────── Direct HTTP API ─────────────────────────┘

The UI communicates with Vault through two paths:

  1. Direct HTTP API — Traditional CRUD operations via the Vault REST API
  2. MCP Proxy — AI-assisted operations through a Node.js proxy using Anthropic's API

Technology Stack

LayerTechnology
FrontendReact 18, TypeScript, Vite
StateZustand, TanStack Query
StylingTailwind CSS, Radix UI
AI IntegrationAnthropic SDK, MCP Proxy
DeploymentDocker, nginx, Kubernetes

Quick Start

Prerequisites

  • Docker 24.0+ and Docker Compose 2.20+
  • HashiCorp Vault 1.12+ (accessible via HTTPS)

Installation

# Clone the repository
git clone https://github.com/linuxus/vault-ai.git
cd vault-ai

# Configure environment
cp .env.example .env
# Edit .env with your Vault address and Anthropic API key

# Build and run
docker compose up -d

# Access the UI
open http://localhost:3000

Project Structure

vault-ai/
├── src/
│   ├── components/          # Reusable UI components
│   ├── features/            # Feature modules
│   │   ├── auth/            # Authentication
│   │   ├── chat/            # AI chat assistant
│   │   └── secrets/         # Secrets management
│   ├── hooks/               # Custom React hooks
│   ├── services/            # API clients (Vault, MCP)
│   ├── stores/              # Zustand state management
│   ├── types/               # TypeScript definitions
│   └── utils/               # Utility functions
├── mcp-proxy/               # MCP proxy server (Node.js)
├── scripts/                 # Setup and demo scripts
├── docs/                    # Documentation
├── docker-compose.yml
├── Dockerfile
└── nginx.conf

Development

# Install dependencies
npm install

# Start development server
npm run dev

# Run tests
npm test

# Build for production
npm run build

Environment Variables

VariableDescriptionRequired
VAULT_ADDRVault server URLYes
ANTHROPIC_API_KEYAnthropic API key for AI featuresYes
VAULT_CACERTPath to the CA bundle that signed Vault's listener cert (for private/self-signed CAs, e.g. vault server -dev-tls)No
VAULT_CAPATHPath to a directory of PEM CA files (alternative to VAULT_CACERT)No
VAULT_SKIP_VERIFYSkip TLS verification (dev only; prefer VAULT_CACERT)No
VAULT_CLIENT_CERT / VAULT_CLIENT_KEYClient certificate + key for mutual TLS to VaultNo
VAULT_TLS_SERVER_NAMEOverride the SNI / certificate hostname during the TLS handshakeNo
VITE_APP_TITLEBrowser tab titleNo

TLS variables are read once at MCP proxy startup — restart the proxy after changing them. A misconfigured VAULT_CACERT/VAULT_CAPATH path fails fast at boot with a clear error.

Documentation

DocumentDescription
Technical SpecArchitecture and implementation details
Design SystemUI components and styling guidelines
MCP IntegrationAI assistant and MCP proxy integration
Deployment GuideDocker deployment
Kubernetes GuideKubernetes deployment with Kind

Security

  • Vault tokens stored in sessionStorage only — never persisted to disk
  • TLS required for Vault connections in production
  • Content Security Policy headers enabled
  • No secret caching in browser storage

Contributing

  1. Fork the repository
  2. Create a feature branch (git checkout -b feature/amazing-feature)
  3. Commit changes (git commit -m 'Add amazing feature')
  4. Push to branch (git push origin feature/amazing-feature)
  5. Open a Pull Request

License

This project is licensed under the MIT License — see the LICENSE file for details.

Acknowledgments

Ver o original no GitHub

Bom para

Relacionados