Jina AI MCP Server

Jina AI MCP server provides AI-powered web search and retrieval, enabling reading page content, extracting structured data, and grounding AI responses with real-time information.

May 4, 2025
GitHub stars

Overview

The Jina AI MCP Server connects AI assistants to Jina's search and retrieval platform. It enables reading web page content, extracting structured data, searching academic papers, and grounding AI responses with real-time information from the web.

Community Server:

Available via Smithery and jina.ai

Key Features

🌐

Web Search

AI-powered web search with configurable result counts and domain filtering

📖

URL Content Reading

Read and extract clean content from any URL with markdown output

📚

Academic Search

Search arXiv, SSRN, and BibTeX databases for academic papers

Parallel Processing

Run multiple search and read operations in parallel for faster results

🖼️

Screenshot Capture

Capture visual screenshots of web pages for context

🏷️

Text Classification

Classify and categorize text content automatically

Available Tools

Quick Reference

ToolPurposeCategory
search_webSearch the web with AI-powered rankingSearch
read_urlExtract clean content from a URLContent
primerGround AI with context before responsesGrounding
parallel_search_webRun multiple searches in parallelSearch
parallel_read_urlRead multiple URLs simultaneouslyContent
search_arxivSearch academic papers on arXivAcademic
capture_screenshot_urlTake a screenshot of a web pageVisual

Detailed Usage

search_web

Search the web using Jina's AI-powered ranking for highly relevant results.

{
  "query": "MCP server best practices production deployment",
  "num_results": 10
}
read_url

Read and extract clean markdown content from any URL.

{
  "url": "https://modelcontextprotocol.io/docs",
  "format": "markdown"
}
search_arxiv

Search academic papers on arXiv by topic, author, or keyword.

{
  "query": "retrieval augmented generation survey",
  "num_results": 10
}
primer

Ground AI responses with real-time context before generating final output.

{
  "topic": "GitHub Copilot MCP integration",
  "num_sources": 5
}

Installation

npx -y @smithery/cli install jina --client claude

API Access:

Available as a managed service via Smithery. Visit smithery.ai/server/jina for connection details.

Sources