MCP.so
登录

Orbis Research MCP Server

@OrbisAPI

基本信息

分类

数据与分析

传输方式

stdio

作者

OrbisAPI

提交者

Orbis API

关于 Orbis Research MCP Server

Access 8,000+ web research and data extraction APIs through a single MCP. Search the web, scrape content, query Reddit, pull Google results, extract structured data, and more. Pay-per-call in USDC via x402 on Base — no API keys, no subscriptions.

Orbis Research MCP Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "orbis-research": {
      "url": "https://orbisapi.com/api/mcp/research"
    }
  }
}

Orbis Research MCP Server 工具

2

Search Research & Data Extraction APIs on the Orbis marketplace. Returns only Research & Data Extraction-specific tools with x402 USDC pay-per-call pricing on Base — no account or signup needed. Try searching for "web search".

Call any API on the Orbis marketplace with x402 USDC payment on Base. Pass x_payment with a signed USDC payment header from x402-fetch (npm) or the x402 Python library.

概览

What is Orbis Research Mcp?

A collection of real web search, scraping, and data extraction APIs for AI agents. It provides access to over a dozen APIs including web search, Reddit search, Google SERP/Places, content classification, and structured data extraction — all billed per call in USDC on Base with no API keys, accounts, or subscriptions required.

How to use Orbis Research Mcp?

Configure the server in your MCP client by adding "url": "https://orbisapi.com/api/mcp/research" under an orbis-research entry in mcpServers. Supported clients include Claude Desktop, Claude Code, Cursor, Windsurf, Cline, and OpenAI Codex CLI. Once connected, prompt the agent with search or scraping requests.

Key features of Orbis Research Mcp

  • Web search returning structured keyword results
  • Reddit post and comment search across subreddits
  • Google Search and Google Maps & Places data
  • Intelligent structured data extraction from web pages
  • Web scraping with rendering and planning tools
  • Pay-per-call USDC billing with no subscription

Use cases of Orbis Research Mcp

  • Search the web for latest news on any topic
  • Research Reddit discussions for market or community insights
  • Extract structured pricing or product data from e-commerce pages
  • Get Google Maps reviews and location data for businesses
  • Generate a scraping strategy for any target website

常见问题

1

What data sources does Orbis Research Mcp access?

It provides real web search, Reddit subreddit search and posts/comments, Google Search results, Google Maps places and reviews, web page content classification, structured data extraction, web scraping, and scraping ROI estimation.

2

Do I need an API key or account to use it?

No. The server requires no API key, no account, and no subscription. Calls are billed per-use automatically.

3

How is billing handled?

Each API call is billed per-use in USDC via x402 on the Base mainnet. Payment is deducted from your wallet as you make requests.

4

Which MCP clients are supported?

Claude Desktop, Claude Code, Cursor, Windsurf, Cline, and OpenAI Codex CLI are explicitly listed. Direct HTTP and x402 usage via Node.js is also documented.

5

Where does the server run and how is data delivered?

The server runs at `https://orbisapi.com/api/mcp/research` as a StreamableHTTP endpoint. Data is fetched in real time from live web sources, search engines, and social platforms.

评论

数据与分析 分类下的更多 MCP 服务器