Jina AI Remote

You run a remote MCP server from Jina that gives you access to Reader, Embeddings, and Reranker APIs through convenient MCP endpoints. It’s designed to be easy to connect to from MCP clients, supports both HTTP-based remote access and local stdio proxies, and works with optional API keys for higher rate limits and better performance.

How to use

To connect your MCP-enabled client to the Jina remote MCP server, choose the preferred connection method and configure your client accordingly. You can either use the remote HTTP endpoint or run a local stdio proxy that forwards to the remote MCP server. Both options let you access a suite of tools for web content, search, and embeddings.

Option 1: HTTP remote MCP endpoint (recommended if your client supports direct remote connections). You’ll point your client to the remote SSE URL and optionally provide an API key in the Authorization header.

Option 2: Local stdio proxy (used when your client does not yet support remote MCP servers). You’ll run a local proxy that forwards to the remote MCP SSE URL and passes your API key in the headers.

How to install

Prerequisites: you need Node.js and npm installed on your system. You’ll also need an API key if you want higher rate limits and access to paid features.

Install and run the HTTP remote MCP server config by connecting your client directly to the remote SSE URL.

Security and keys

Some tools require a Jina API key for higher rate limits and broader access. You can obtain a free API key from jina.ai and pass it in the Authorization header as a Bearer token when you configure your MCP client.

Troubleshooting

If you encounter issues with tool calling loops, content truncation, or disconnections, first verify that your API key is valid and that your MCP client supports the chosen connection method. If a client shows a UI error while using a local proxy, try restarting the proxy to refresh tool definitions.

Developer notes

The server exposes a wide set of tools for web content extraction, search, and embeddings. When you integrate, consider using parallel tool variants for faster results and combining URL reads with searches to ensure you get sourced content alongside structured results.

Available connection configurations

{
  "mcpServers": {
    "jina-mcp-server": {
      "url": "https://mcp.jina.ai/sse",
      "headers": {
        "Authorization": "Bearer ${JINA_API_KEY}"
      }
    }
  }
}

{
  "mcpServers": {
    "jina-mcp-server": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://mcp.jina.ai/sse",
        "--header",
        "Authorization: Bearer ${JINA_API_KEY}"
      ]
    }
  }
}

Tools and capabilities

The MCP server provides a suite of tools to fetch, parse, and search content, including contextual info, web/text extraction, and various search endpoints. You can enable/disable tools as needed in your client configuration and combine multiple tools to build rich conversational flows.

Available tools

primer

Get current contextual information for localized, time-aware responses

read_url

Extract clean, structured content from web pages as markdown via Reader API (optional API key)

capture_screenshot_url

Capture high-quality screenshots of web pages via Reader API (optional API key)

guess_datetime_url

Analyze web pages for last update/publish datetime with confidence scores

search_web

Search the entire web for current information and news via Reader API (requires API key)

search_arxiv

Search academic papers on arXiv via Reader API (requires API key)

search_ssrn

Search SSRN papers via Reader API (requires API key)

search_images

Search images across the web via Reader API (requires API key)

expand_query

Expand and rewrite search queries using the query expansion model via Reader API (requires API key)

parallel_read_url

Read multiple web pages in parallel for efficient content extraction via Reader API (optional)

parallel_search_web

Run multiple web searches in parallel for broad topic coverage via Reader API (requires API key)

parallel_search_arxiv

Run multiple arXiv searches in parallel for comprehensive coverage (requires API key)

parallel_search_ssrn

Run multiple SSRN searches in parallel for social science coverage (requires API key)

sort_by_relevance

Rerank documents by relevance to a query via Reranker API (requires API key)

deduplicate_strings

Find top-k semantically unique strings via Embeddings API and submodular optimization (requires API key)

deduplicate_images

Find top-k semantically unique images via Embeddings API and submodular optimization (requires API key)