Jina AI
The official Jina AI MCP server gives AI agents a toolkit for reading and searching the web. Agents can read URLs to clean markdown, capture screenshots, search the web, arXiv, images, and academic papers, expand queries, rerank and deduplicate results, classify text, and read or search many pages in parallel.
Tools
capture_screenshot_url
Capture a high-quality screenshot of a given web page.
classify_text
Classify texts into user-defined labels via embeddings.
deduplicate_strings
Return the top-k semantically unique strings from a set.
expand_query
Expand and rewrite a search query for better recall.
guess_datetime_url
Analyze a page to identify its publish or update date.
parallel_read_url
Read multiple web pages simultaneously for efficiency.
parallel_search_web
Run multiple web searches in parallel for coverage.
primer
Get current contextual info for time-aware responses.
read_url
Extract clean, structured markdown content from a URL.
search_arxiv
Search academic papers and preprints hosted on arXiv.
search_bibtex
Return academic papers together with their BibTeX citations.
search_images
Search for images across the web using a text query.
search_web
Search the entire web for current information and news.
sort_by_relevance
Rerank a set of documents by relevance to a query.