# Tavily

Tavily gives an agent web search, page extraction, site mapping, and multi-source research. Use it when answers need current facts or documentation.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · 1575 installs
- Author: tavily
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://smithery.ai/server/Tavily
- Updated: 2026-03-27
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/mcp/tavily

## What it does

Tavily adds HTTP tools for finding current web information, reading content from a URL, crawling a site, and listing its discovered pages. `tavily_search` returns snippets with source URLs, while `tavily_research` builds an answer from material gathered across multiple sources.

The `tavily_skill` tool searches documentation for a library, API, or other tool and returns query-specific structured chunks. The server uses `https://tavily.run.tools`; in Claude Code, add it with `claude mcp add --transport http tavily https://tavily.run.tools`. This fits tasks where source links matter, because the search result includes them.

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

## Install

### Claude Code (native platform)

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

### Codex

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

### Cursor

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

### GitHub Copilot

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

### Gemini CLI

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

## When to use it
- find current information on the web
- extract page content from a URL
- crawl a site from its starting URL
- map the URLs discovered on a website
- search documentation for a library or API

## Example prompts
- Find current information about this library and include the source URLs.
- Extract the page at this URL and return its content as Markdown.
- Research `FastAPI` documentation and explain how to build this endpoint.

## Similar
- [web-search-tavily](https://affpapa.org/en/nevibecode/skills/web-search-tavily.md) — skill · 1100 installs
- [tavily-web](https://affpapa.org/en/nevibecode/skills/tavily-web.md) — skill · 94 installs
- [tavily-search](https://affpapa.org/en/nevibecode/skills/tavily-search.md) — skill · 39400 installs
- [tavily-best-practices](https://affpapa.org/en/nevibecode/skills/tavily-best-practices.md) — skill · 21100 installs
- [tavily-usage](https://affpapa.org/en/nevibecode/skills/tavily-usage.md) — skill · 900 installs
- [tavily-tools](https://affpapa.org/en/nevibecode/plugins/tavily-tools.md) — plugin · ★ 1216 repo stars

---
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
