# Tavily MCP Extract

> Extract clean markdown or text content from specific URLs through the local Tavily MCP server. Use this skill when the user provides one or more URLs and wants their content, says "extract", "grab the content from", "pull the text from", "get the page at", "read this webpage", or needs clean text from web pages. Supports advanced extraction for JavaScript-heavy pages and query-focused extraction.

- Skill: `minhngoc25a/tavily-mcp-extract` (Agent Skill)
- Install (CLI): `npx skillmds@latest add minhngoc25a/tavily-mcp-extract`
- Raw SKILL.md: https://api.skillmd.com/api/skills/minhngoc25a/tavily-mcp-extract/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: minhngoc25a (https://skillmd.com/u/minhngoc25a)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/minhngoc25a/tavily-mcp-extract

---


# Tavily MCP extract

Use `mcp__tavily-mcp__tavily_extract` to read known URLs and return clean page content.

## When to use

- The user provides specific URLs.
- Search or map found URLs and you need page content.
- You need clean markdown or text from webpages, including JavaScript-heavy pages.

For URL discovery on a site, use `tavily-mcp-map`. For many pages on a site, use `tavily-mcp-crawl`.

## Parameters

| Parameter | Use |
| --- | --- |
| `urls` | List of URLs to extract. Batch related URLs together. |
| `extract_depth` | `basic` for normal pages, `advanced` for JavaScript-heavy pages, tables, or embedded content. |
| `format` | `markdown` by default, or `text` for plain text. |
| `query` | Rerank or focus returned chunks around a specific question or topic. |
| `include_images` | Include image URLs from pages. |
| `include_favicon` | Include favicon URLs. |

## Depth selection

| Depth | Best for |
| --- | --- |
| `basic` | Most static pages and faster extraction. |
| `advanced` | Dynamic pages, JavaScript-rendered apps, tables, and pages where basic extraction misses content. |

## Patterns

### Read one URL

Call extract with a single-item `urls` list and default markdown format.

### Read several related URLs

Batch URLs in one call when they are part of the same task.

### Targeted extraction

Use `query` when the page is long and only topic-relevant chunks are needed.

### Retry missing content

If `basic` returns incomplete content for a page that likely renders dynamically, retry with `extract_depth: advanced`.

## Tips

- Use search or map first when the exact URL is unknown.
- Prefer `markdown` for structured pages and documentation.
- Prefer `text` when the user needs plain prose or downstream processing.
- Do not use crawl for one or two known URLs; extract is more direct.