Skip to main content

🕷️ Arachne MCP Server

9 MCP tools para web scraping, browser automation, visão computacional, transcrição de áudio e RAG — tudo via API.

Conecta no Arachne como backend. Funciona com Claude Desktop, Cursor, Codex CLI, e qualquer cliente MCP.

✨ Tools

Tool O que faz Ideal para
arachne_search Busca na web via DuckDuckGo Pesquisa, coleta de informações
arachne_scrape Extrai markdown limpo de URLs Páginas estáticas, blogs, docs
arachne_extract Extrai QUALQUER formato (áudio, vídeo, PDF, YouTube) Canivete suíço
arachne_browser_extract Navegador real com evasão Cloudflare/CAPTCHA Sites que bloqueiam scraper
arachne_browser_run Executa ações em navegador (click, type, login) Automação de formulários
arachne_query Pergunta pra sua base de conhecimento RAG Chatbot com seus dados
arachne_vision Analisa imagens: OCR, cores, faces, descrição AI Extrair texto de fotos
arachne_transcribe Transcreve áudio/vídeo/YouTube com Whisper Podcast, reunião, vídeo
arachne_capabilities Auto-descoberta de capacidades Saber o que o Arachne faz

🚀 Quick Start

1. Pegue uma API Key

Crie em arachne.seu.pet/dev (plano Free: 500 req/mês).

2. Configure no Claude Desktop

{
  "mcpServers": {
    "arachne": {
      "command": "python3",
      "args": ["-m", "arachne_mcp"],
      "env": {
        "ARACHNE_API_KEY": "sua_chave_aqui"
      }
    }
  }
}

3. Ou teste direto

export ARACHNE_API_KEY="sua_chave"
python3 arachne_mcp.py

Precisa de httpx: pip install httpx

📦 Como funciona

O MCP server é um cliente HTTP que chama a API pública do Arachne. Zero dependência de infra local — roda de qualquer lugar.

Seu agente AI → MCP stdio → arachne_mcp.py → HTTP → Arachne API → resultado

🔧 Exemplos

# Via arachne-sdk (pip install arachne-sdk)
from arachne_sdk import Arachne
client = Arachne(api_key="sua_chave")

# Busca na web
results = client.search("preço iPhone 16 Brasil")
for r in results:
    print(f"{r.title}: {r.url}")

# Extrair página
content = client.scrape("https://exemplo.com")
print(content[:500])

# Analisar imagem
vision = client.vision("https://exemplo.com/foto.jpg")
print(vision.ocr.text)

# Perguntar ao RAG
answer = client.query("Qual a diferença do Arachne pro Firecrawl?")
print(answer)

📊 Planos

Plano Preço Requests/mês Features
Free R$ 0 500 search, scrape, jobs
Pro R$ 49/mês 10.000 + browser, vision, transcribe, MCP
Enterprise R$ 199/mês 100.000 + admin, export, suporte dedicado

🏗️ Stack

  • Backend: FastAPI + Crawl4AI + Whisper + Tesseract + SQLite
  • Engines: Trafilatura → Crawl4AI SDK → Sidecar Docker → Camoufox
  • MCP Transport: stdio (compatível com Claude Desktop, Cursor, Codex)

🔗 Links


🕷️ Built with the Arachne engine. Open source MCP server.

Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

arachne_mcp-1.0.1.tar.gz (6.8 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

arachne_mcp-1.0.1-py3-none-any.whl (7.7 kB view details)

Uploaded Python 3

File details

Details for the file arachne_mcp-1.0.1.tar.gz.

File metadata

  • Download URL: arachne_mcp-1.0.1.tar.gz
  • Upload date:
  • Size: 6.8 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.15

File hashes

Hashes for arachne_mcp-1.0.1.tar.gz
Algorithm Hash digest
SHA256 59a06e670edca6ace6ec74f85c360b8b56646e5d094d73d3ac22fa935dc1657e
MD5 5d49e83214ffbaea8ad98e067f8c7835
BLAKE2b-256 3ae19091e9a8bfb932f7e15e6ee2ff105e15e6e873e462c21c1d1b07c89f6257

See more details on using hashes here.

File details

Details for the file arachne_mcp-1.0.1-py3-none-any.whl.

File metadata

  • Download URL: arachne_mcp-1.0.1-py3-none-any.whl
  • Upload date:
  • Size: 7.7 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.11.15

File hashes

Hashes for arachne_mcp-1.0.1-py3-none-any.whl
Algorithm Hash digest
SHA256 32b4eb3639fa54628bca08aebbad1817839186b7367199eb4ddc019ff349e19d
MD5 5f260696afcaac6bb5c8f0a158ae13e5
BLAKE2b-256 14f35eb13c51eed4af569b140c71b49a43134adb4a6dd057f27923f654f676fb

See more details on using hashes here.

Release history Release notifications | RSS feed

1.4.0

2 files

1.3.0

2 files

1.2.0

1 file

1.1.2

1 file

1.1.1

1 file

1.1.0

1 file

This release

1.0.1 This release

2 files

1.0.0

2 files

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page