# Exa Search

Exa Search searches the web and converts live web pages into clean markdown for AI assistants. Helpful when fetching fresh documentation for third-party libraries and APIs.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 9181 installs
- Author: exa
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/exa
- Updated: 2024-12-13
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/exa-search

## What it does

The server runs over HTTP using the remote endpoint `https://exa.run.tools`. It exposes two tools: `web_search_exa` and `web_fetch_exa`. The search tool looks up topics based on descriptions of the target content rather than plain keyword lists. It returns cleaned text excerpts.

The fetch tool converts target pages into plain markdown. It handles cases where initial snippets are not enough or when specific web addresses are already known. You can pass several addresses in one batch to extract page text and metadata together. Because this setup relies on an external HTTP endpoint, the agent must be able to reach `https://exa.run.tools`.

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

## Install

### Claude Code (native platform)

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

### Codex

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

### Cursor

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

### GitHub Copilot

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

### Gemini CLI

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

## When to use it
- search for fresh documentation on developer libraries and APIs
- extract clean markdown text from web pages
- retrieve content and metadata from multiple links in one batch
- look up current facts, news, and company details

## Example prompts
- Use `web_search_exa` to find recent changes in the library documentation.
- Fetch the full markdown content of the provided link with `web_fetch_exa`.
- Search for technical articles comparing state management solutions.

## Similar
- [exa-search](https://affpapa.org/en/nevibecode/skills/affaan-m-exa-search.md) — skill · 6300 installs
- [exa-web-search-free](https://affpapa.org/en/nevibecode/skills/exa-web-search-free.md) — skill · 2800 installs
- [Keenable Web Search](https://affpapa.org/en/nevibecode/mcp/keenable-web-search.md) — MCP server · 2498 installs
- [exa](https://affpapa.org/en/nevibecode/plugins/exa.md) — plugin · ★ 5254 repo stars
- [exa-contents](https://affpapa.org/en/nevibecode/skills/exa-contents.md) — skill · 1400 installs
- [cli-anything-exa](https://affpapa.org/en/nevibecode/skills/cli-anything-exa.md) — skill · 555 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
