# Jina AI

The Jina AI MCP server gives an agent web search, page reading, and PDF extraction. Use it when research needs current sources or structured results.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 3481 installs
- Author: jina
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/jina
- Updated: 2026-03-29
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/jina-ai

## What it does

Jina AI adds web retrieval to an agent. It can search the web, turn page content into readable markdown, inspect several URLs in parallel, and expand an initial query into broader searches. Separate tools cover arXiv, SSRN, Jina AI blog posts, and image search.

The server also classifies text, reranks documents, selects semantically distinct strings or images, finds papers with BibTeX output, and extracts figures, tables, and equations from PDFs. It connects through `https://jina.run.tools` and is listed for Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, and Gemini CLI.

\`show_api_key` can return the bearer token held in the MCP authorization header, so access to that tool should be controlled.

## Requirements
- Remote server: `https://jina.run.tools`

## Install

### Claude Code (native platform)

**Platform CLI: remote server**
```bash
claude mcp add --transport http jina https://jina.run.tools
```

### Codex

**Platform CLI: remote server**
```bash
codex mcp add jina-ai --url https://jina.run.tools
```

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "jina-ai": {
            "url": "https://jina.run.tools"
        }
    }
}
```

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "jina-ai": {
            "type": "http",
            "url": "https://jina.run.tools"
        }
    }
}
```

### Gemini CLI

**Platform CLI: remote server**
```bash
gemini mcp add -t http jina-ai https://jina.run.tools
```

## When to use it
- search the web for current sources
- compare content from several URLs
- find papers and produce BibTeX citations
- extract figures, tables, and equations from a PDF
- rerank documents against a query

## Example prompts
- Search the web for current sources on this topic and compare the findings.
- Read `URL_1` and `URL_2`, then list the points they share and where they differ.
- Extract figures, tables, and equations from this PDF, then return BibTeX entries for the papers you find.

## Similar
- [jina-reader](https://affpapa.org/en/nevibecode/skills/jina-reader.md) — skill · 1000 installs
- [arXiv](https://affpapa.org/en/nevibecode/mcp/arxiv.md) — MCP server · 5142 installs
- [Tavily](https://affpapa.org/en/nevibecode/mcp/tavily.md) — MCP server · 1575 installs
- [paper-search-mcp-openai-v2](https://affpapa.org/en/nevibecode/mcp/paper-search-mcp-openai-v2.md) — MCP server · 7042 installs
- [Brave Search](https://affpapa.org/en/nevibecode/mcp/brave-search.md) — MCP server · 87579 installs
- [literature-search-arxiv](https://affpapa.org/en/nevibecode/skills/literature-search-arxiv.md) — skill · 3700 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md
