MCP tool reference · #135 of 563 by publishers
extract
Exposed by 20 servers from 17 publishers, read from the live servers by the Serverdex check; most of them sit in Search & Web; 65% of them take a url parameter; the most-starred is SerpApi (186 stars).
extract(url: string, format?: string, schema: object, …)Parameters
share = % of the 20 servers whose schema has it · required = % of those that mark it required| Parameter | Type | Share | Required | Usual description |
|---|---|---|---|---|
| url | string | 65% | 84.6% | Absolute http(s) URL of the page to fetch |
| format | string | 35% | 0% | Returned representation: text (default), markdown, or both. HTML structured details are included in every format; Markdown/plain sources have empty details. |
| schema | object | 25% | 100% | Field name -> type or description |
| maxChars | integer | 20% | 0% | Maximum characters of body text to return (500-100000, default 20000) |
| section | string | 20% | 0% | Optional visible HTML ID within primary content, 1-500 characters. A leading # (the fragment a docs link shows) or a URL ending in #id is accepted, and so is a |
| urls | array | 15% | 66.7% | A single URL string, or an array of URL strings, to extract clean text from. |
| mode | string | 10% | 0% | Extraction mode: auto (extract=auto, default), css (requires css_extractor), or the deprecated autoparse |
| text | string | 10% | 50% | Messy text to turn into JSON (up to 40,000 characters) |
| wait | number | 10% | 0% | Milliseconds to wait after load. Requires js_render=true; without it, including in automatic mode, it may be ignored. |
| _apiKey | string | 5% | 0% | Optional, your own Tavily API key for higher limits; omit to use the shared Pipeworx key. |
| api_key | — | 5% | 0% | — |
| bars | — | 5% | 0% | Bar range such as "1-8", "5" or "9-" |
Source: the input schema each server returns for extract, aggregated over the 20 index-tier servers; type and description are the values most servers use. Rebuilt every 6 hours.
Servers with extract
19 of 20 · 17 publishers · 19 answering live · most-starred first, 3 per publisher| # | Server | Publisher | Category | ★ | Live | How it describes extract |
|---|---|---|---|---|---|---|
| 1 | SerpApiserpapi · Search & Web | serpapi | Search & Web | 186 | yes | Extract content from a public http(s) URL, discover site URLs, or crawl pages. API options go in params: url is required; params.mode is scrape (default), map, or crawl. See serpapi://engines/extract for all options. Top-level mode controls the response: - complete (default): full JSON. - compact: native Markdown by default. params.output selects the format and overrides either default: - "md": |
| 2 | Geonode MCP Servergeonodecom · Search & Web | geonodecom | Search & Web | 27 | yes | Extract content from a web page. Supports HTML and Markdown output formats, optional JavaScript rendering via headless browser, proxy geo-targeting, custom HTTP headers, and explicit browser wait controls via the wait_config object. Can run synchronously (returns content immediately) or asynchronously (returns a job ID that can be polled with the 'job' tool). Use 'render_js=true' for pages that re |
| 3 | Zenrows MCPzenrows · Search & Web | zenrows | Search & Web | 21 | yes | Extract structured data from a webpage via Zenrows. Returns JSON fields (products, articles, listings) in "data" rather than a full page body. Set include_html to also get the page HTML. Modes: - auto (default): extract=auto, site-tailored Extract (currently free, billing may apply later). Works on any prepared domain, with any API key. Error REQS007 says why in "reason": - domain_not_prepa |
| 4 | Agent Data Tools: Web, documents and searchnutshell2000 · Search & Web | nutshell2000 | Search & Web | 1 | yes | Fetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context. Static HTML, Markdown or text; optional section ID. Default 20,000 output characters, max 100,000; input 2 MiB. No JavaScript rendering (see /extract/rendered), logins or paywalls. Empty pages, bad sections, browser challenges and errors are not |
| 5 | Agent Data Tools: Documents to textnutshell2000 · Documents & Knowledge | nutshell2000 | Documents & Knowledge | 1 | yes | Fetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context. Static HTML, Markdown or text; optional section ID. Default 20,000 output characters, max 100,000; input 2 MiB. No JavaScript rendering (see /extract/rendered), logins or paywalls. Empty pages, bad sections, browser challenges and errors are not |
| 6 | Agent Data Tools: Full list (every tool as its own definition)nutshell2000 · Documents & Knowledge | nutshell2000 | Documents & Knowledge | 1 | yes | Fetch any public web page URL and get clean, LLM-ready text or Markdown: main content, code blocks, tables, links and headings, for RAG and agent context. Static HTML, Markdown or text; optional section ID. Default 20,000 output characters, max 100,000; input 2 MiB. No JavaScript rendering (see /extract/rendered), logins or paywalls. Empty pages, bad sections, browser challenges and errors are not |
| 7 | Desearchdesearch-ai · Communication | desearch-ai | Communication | 1 | yes | Extract a public URL and return its content as plain text, markdown or HTML using Desearch. Markdown keeps headings, lists and tables. Text is the page's main content; set full_page for every section, such as opening hours or fee tables. A js_required error means the page needs js: true; blocked means the site refused or showed a bot check. Preferred over web-crawl for new integrations. API docs: |
| 8 | x402 Data Gatewayepistemedeus · Crypto & Web3 | epistemedeus | Crypto & Web3 | 1 | yes | Extract metadata, Open Graph/Twitter tags, JSON-LD, headings, links, a bounded excerpt, and AI-readiness flags from a public HTTP(S) page. Set textExcerptLimitChars to adjust the excerpt. Optional integer from 1 to 40000 (default 1200). Invalid values are rejected before payment. Price unchanged. Use `read` instead for Markdown. Follows redirects with SSRF, timeout, and size guards; no JavaScript. |
| 9 | Nightshift agent tools (run by an AI)llm-9-to-5 · Search & Web | llm-9-to-5 | Search & Web | — | yes | Fetch any public URL and return clean markdown of the main content, plus extracted links and page metadata. Built for LLM agents. (Free via MCP, 50 calls/day; unlimited pay-per-call over x402 at $0.002: /api/extract) |
| 10 | Helvehelve-dev · APIs & Integrations | helve-dev | APIs & Integrations | — | yes | Extract clean, readable content from one or more web page URLs (articles, docs, listings). Pass query to pull only the sections relevant to a topic instead of the whole page. Content is markdown. |
| 11 | Tavily (Unofficial)pipeworx-io · Search & Web | pipeworx-io | Search & Web | — | yes | Extract clean article text from one or more URLs via Tavily: strips boilerplate/navigation and returns up to 20,000 chars of readable content per page. Accepts a single URL string or an array. Ideal for feeding source pages into an LLM. |
| 12 | SaltCrate Oraclesaltcrate · Search & Web | saltcrate | Search & Web | — | yes | Schema-driven structured extraction: fetch a URL and map its content into typed entities that conform to a JSON Schema you provide. Pass the schema as a JSON string; returns schema-conforming JSON (e.g. pricing, company facts, product specs, events). Requires DEEPSEEK_API_KEY. |
| 13 | Tolcayayabough · Search & Web | yayabough | Search & Web | — | yes | Extract structured data from messy text into clean JSON. Input: text, schema (JSON object describing desired fields). Costs $0.02 USDC per call (x402 v2, Base or Solana). Send the HTTP header PAYMENT-SIGNATURE with your x402 payment, or call POST https://tolca.tonbot.ca/v1/extract with an x402 client. |
| 14 | Scoreproofasampat3090 · Media & Creative | asampat3090 | Media & Creative | — | yes | Save chosen parts and/or bars of a score as new files. |
| 15 | PrepLayer.aimaximusboomerus-afk · Other | maximusboomerus-afk | Other | — | yes | Extract structured fields from unstructured text with byte offsets and normalized values: emails, URLs, IPv4, phones, money amounts, ISO dates, UUIDs, EVM addresses, SKUs, semver. $0.005 USDC per successful call (x402, client-paid); failed calls are not charged. |
| 16 | срезAI — MCP-сервер и поисковый API для ИИ-агентовsrezai-team · Search & Web | srezai-team | Search & Web | — | yes | Читает страницу и возвращает JSON строго по вашей схеме: только запрошенные поля, без текста страницы. Чего на странице нет - приходит null, значения не домысливаются. Когда: нужны 2-5 конкретных значений - цена, характеристики, автор и дата, список вакансий. Экономит контекст: вместо всей страницы придут только поля. Когда не: нужен связный текст или вы не знаете заранее, что искать, - read_url |
| 17 | SiteCheckbck-stack · Crypto & Web3 | bck-stack | Crypto & Web3 | — | yes | Structured data from a web page (url) or your own text: give a schema of up to 20 field names with what each holds, and get a JSON object with exactly those fields, null where the text does not state one. Llama 3.3 70B. Text is capped at 12,000 characters; a model answer that is not JSON is retried once, then not charged. Price: $0.004 per successful call, paid with x402 (USDC on Base, Solana and |
| 18 | Mercury x402 MCPmercuryx402 · Search & Web | mercuryx402 | Search & Web | — | yes | TYPED structured extract for autonomous agents — URL + schema → a clean, type-safe JSON record. Where /buy/fetch returns page TEXT (and ?extract= returns string-only fields), THIS returns the schema-conformant object an LLM/RAG/trading pipeline actually consumes: pass ?url=…&schema=title,price:number,rating:number,inStock:boolean and get back { title:"…", price:19.99, rating:4.5, inStock:true } — |
| 19 | CleanTextwminno · Search & Web | wminno | Search & Web | — | yes | Fetch a public web page and return its main article text or Markdown (nav, ads and boilerplate stripped). |
Stars from GitHub as stored in the Serverdex index (re-read every 7 days per server); live = answered the last daily check of its hosted endpoint. The other 1 servers are reached from their publisher pages.