convez.ai
MCP ServerIntegração & APIsIntermediário
junipr-labs

Junipr MCP Server

junipr-labs/mcp-server

Web intelligence tools: screenshots, PDFs, metadata, scrapers e mais

Ver no GitHub
InicialPluginScore · de 0 a 100

Servidor MCP que integra as APIs Junipr (75+ ferramentas de web intelligence) ao Claude e outros clientes MCP. Oferece captura de screenshots, geração de PDFs, extração de metadados e acesso a scrapers, validadores e conversores. Ideal para automação de tarefas web e análise de páginas.

Instalação

$npx convez-ai junipr-labs/mcp-server

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
50%
Último commit
há 8 meses
Versão
v1.2.0
Issues abertas
0
Contribuidores
1

Casos de uso

  • Capturar screenshots de páginas web para análise visual
  • Gerar PDFs de URLs ou HTML dinamicamente
  • Extrair metadados, Open Graph e JSON-LD de websites
  • Automatizar scraping e validação de dados web
  • Analisar e converter conteúdo web em diferentes formatos

Veredito

MCP server bem estruturado e prático que democratiza acesso a 75+ ferramentas Junipr via interface padrão. Excelente para quem precisa automação web integrada ao fluxo de IA, com setup simples e plano gratuito generoso (500 créditos/mês).

README

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

junipr-mcp

MCP server for Junipr APIs. Gives AI assistants access to 75+ web intelligence tools — screenshots, PDFs, metadata extraction, scrapers, validators, and more.

Tools

ToolDescription
screenshotCapture a screenshot of any webpage (PNG, JPEG, or WebP)
pdfGenerate a PDF from a URL or raw HTML
metadataExtract title, description, Open Graph, Twitter Cards, JSON-LD, and more
run_toolRun any of 75+ Junipr tools by slug (scrapers, validators, converters, analyzers)

Setup

1. Get an API key

Sign up at junipr.io/login to get a free API key (500 credits/month).

2. Configure your MCP client

Claude Desktop

Add to your Claude Desktop config file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "junipr": {
      "command": "npx",
      "args": ["-y", "junipr-mcp"],
      "env": {
        "JUNIPR_API_KEY": "your-api-key-here"
      }
    }
  }
}

Claude Code (CLI)

claude mcp add junipr -- npx -y junipr-mcp

Then set the environment variable:

export JUNIPR_API_KEY=your-api-key-here

Cursor

Add to your Cursor MCP settings (.cursor/mcp.json):

{
  "mcpServers": {
    "junipr": {
      "command": "npx",
      "args": ["-y", "junipr-mcp"],
      "env": {
        "JUNIPR_API_KEY": "your-api-key-here"
      }
    }
  }
}

VS Code

Add to your VS Code settings (.vscode/mcp.json):

{
  "servers": {
    "junipr": {
      "command": "npx",
      "args": ["-y", "junipr-mcp"],
      "env": {
        "JUNIPR_API_KEY": "your-api-key-here"
      }
    }
  }
}

Windsurf

Add to your Windsurf MCP config (~/.codeium/windsurf/mcp_config.json):

{
  "mcpServers": {
    "junipr": {
      "command": "npx",
      "args": ["-y", "junipr-mcp"],
      "env": {
        "JUNIPR_API_KEY": "your-api-key-here"
      }
    }
  }
}

Tool Reference

screenshot

Capture a screenshot of any webpage.

ParameterTypeDefaultDescription
urlstring(required)URL to capture
formatpng | jpeg | webppngImage format
widthnumber1280Viewport width (320-3840)
heightnumber720Viewport height (200-2160)
fullPagebooleanfalseCapture full scrollable page
deviceTypedesktop | mobile | tabletdesktopDevice to emulate
blockCookieBannersbooleantrueHide cookie banners

pdf

Generate a PDF from a URL or HTML.

ParameterTypeDefaultDescription
urlstring—URL to convert (provide url or html)
htmlstring—Raw HTML to convert (provide url or html)
formatA4 | Letter | Legal | Tabloid | A3 | A5A4Paper format
landscapebooleanfalseLandscape orientation
printBackgroundbooleantrueInclude backgrounds
marginobject1cm each{ top, right, bottom, left } in CSS units
headerTemplatestring—Header HTML template
footerTemplatestring—Footer HTML template
displayHeaderFooterbooleanfalseShow header/footer

metadata

Extract metadata from any webpage.

ParameterTypeDefaultDescription
urlstring(required)URL to extract metadata from

Returns title, description, Open Graph tags, Twitter Cards, JSON-LD structured data, favicon, canonical URL, language, and more.

License

MIT

Ver o original no GitHub

Bom para

Relacionados