Exa

Exa MCP Connector for Claude

A+

Semantic search engine built for AI — find conceptually relevant web content, not just keyword matches. Powered by neural search technology.

3 tools Official Updated Jun 28, 2026 Official Vinkius Partner

Connect your AI agent to Exa — the semantic search engine built from the ground up for AI applications.

What you can do

  • Semantic Search — Search the web using natural language. Unlike Google, Exa understands concepts and meaning, returning results that are semantically relevant even without exact keyword matches
  • Find Similar — Provide any URL and discover web pages with similar content. Perfect for competitive analysis, research expansion, and content discovery
  • Extract Contents — Get clean text, highlights, and summaries from any list of URLs. Ideal for building knowledge bases from curated sources

How it works

  1. Subscribe to this server
  2. Enter your Exa API key (1,000 free searches/month)
  3. Your agent can now search the web with semantic understanding

Who is this for?

  • AI Agent Developers — give agents web search that truly understands context and meaning
  • Research Teams — find related papers, articles, and resources that keyword search would miss
  • Content Teams — discover competitors, alternatives, and related content at scale
semantic-searchneural-searchweb-indexinginformation-retrievalai-data-sourcingcontextual-search

3 tools expose this connector's capabilities to your AI agent.

exa_find_similar

Useful for finding competitors, related articles, or alternative sources on the same subject. Find web pages semantically similar to a given URL. Perfect for competitive analysis and content discovery

exa_get_contents

Useful when you already know which pages you want to read and need their content in a structured format. Extract clean text content from specific URLs. Provide comma-separated URLs to retrieve their content

exa_search

Returns page text, highlights, and relevance scores. Supports search types: auto (default), instant (fastest), fast, deep (most thorough). Search the web using Exa semantic search engine. Finds conceptually relevant results, not just keyword matches

See how to talk to your AI agent using Exa.

Search for companies building memory infrastructure for AI agents.

Found 10 semantically relevant results: 1. **Mem0** (mem0.ai) — Score: 0.94 Leading memory layer for AI agents with hybrid KV+vector+graph storage... 2. **Zep** (getzep.com) — Score: 0.91 Long-term conversational memory with temporal knowledge graphs... 3. **Letta** (letta.com) — Score: 0.88 OS-inspired agent runtime with self-managed memory hierarchy...

Find pages similar to https://docs.langchain.com/docs/get_started/introduction

Found 10 pages similar to LangChain's introduction: 1. **LlamaIndex Documentation** (docs.llamaindex.ai) — Score: 0.92 2. **CrewAI Getting Started** (docs.crewai.com) — Score: 0.89 3. **AutoGen Documentation** (microsoft.github.io/autogen) — Score: 0.86 4. **Haystack Introduction** (docs.haystack.deepset.ai) — Score: 0.83

Extract the content from these 3 URLs: https://arxiv.org/abs/2401.00001, https://openai.com/blog, https://anthropic.com/research

Extracted content from 3 URLs: ### 1. arxiv.org — Research Paper > Key highlights: Novel approach to multi-agent coordination using graph neural networks... ### 2. openai.com — Blog Post > Full text extracted: 4,200 words covering latest research developments... ### 3. anthropic.com — Research Page > Highlights: Constitutional AI methodology and safety-first approach...

Exa uses neural embedding models to understand the meaning of your query, not just keywords. When you search 'startups building AI infrastructure', Google returns pages containing those exact words. Exa returns companies that match that concept — even if their pages use completely different terminology. It also provides specialized search indexes for people, companies, code, and academic papers.

Related Connectors