Crawl a website from a seed URL, follow linked pages up to a configurable depth, and scrape each page into markdown or HTML. Use it to extract site content, analyze structure, and gather data across multiple pages.
Server Side Public , v 1
510,057 tools. Updated 2026-09-03 19:32
"How to fetch or scrape data from a website for use in training an LLM" matching MCP tools:
- Turn any URL into clean, LLM-ready markdown or text. Fetch full page content, or get relevance-ranked highlights by passing a query.MIT
- Register a generation model for testing by selecting an LLM from the registry, enabling use with run_test.Apache 2.0
- Search FLOX documentation to find how-to guides, references, and explanations. Get grounded answers about behavior, configuration, or APIs instead of relying on training data.MIT
- Scrape a URL with a pre-created AI parser to extract structured data as JSON. Use when you have a finished parser and need consistent data from similar pages.MIT
- Refresh historical training data for listed tickers from Yahoo Finance or FRED, with full backfill from 2000 for new features and incremental updates to today for existing ones.MIT
Matching MCP Servers
- AlicenseNot gradedqualityCmaintenanceA minimal MCP server with get_weather and create_ticket tools, used for testing MCP servers across protocol, unit, eval, transport, and auth layers.MIT
- AlicenseNot gradedqualityBmaintenanceEnables LLM agents to process local videos into timestamped, citable text documents and then query them through tools for listing videos, retrieving transcripts, and fetching specific segments, all fully offline.MIT
Matching MCP Connectors
Turn any website into structured JSON data matching your custom schema.
India Open Government Data (OGD) Platform MCP — data.gov.in
- Deploy a PHP application to an Agency Plan website by uploading an archive file and extracting it, overwriting existing contents. Use for PHP sites that need no build step.MIT
- Generate a concise 3-5 sentence summary of a webpage or raw text using an LLM. Ideal for agents needing a quick, information-dense digest without running their own fetch or summarization.MIT
- Scrape JavaScript-rendered pages for SEO audit signals: title, meta tags, headings, Open Graph, structured data, canonical, images, and issues. Waits for JS execution before extracting content.MIT
- Fetch a specific Hudu password by its numeric ID. Retrieve credential details for a company asset or website, using the ID from a prior list query.MIT
- Fetch a single website by numeric id to view its live monitoring data: uptime, TLS certificate expiry, WHOIS, and DNS records.MIT
- Batch scrape up to 100 URLs in parallel into LLM-ready markdown, with background job support and options for JavaScript-heavy pages.Unlicense - libtelnet variant
- Import an Agency Plan website from an uploaded archive. After uploading the archive to the website root, provide its filename to apply the archive's contents to the site.MIT
- Lists available PHP versions for websites under an Agency Plan order based on its hosting server. Use before creating a website to choose a compatible PHP version.MIT
- Fetch a Hudu company by numeric ID to get its full record. Use after listing companies to access detailed data for a specific ID.MIT
- Create a dataset from an execution run, linking it for provenance. Use when producing curated collections like augmented training data or filtered results.Apache 2.0
- Fetch a documentation page as clean Markdown, making it ready for LLM consumption. Use with search_docs URLs from approved docs.ansible.com, docs.redhat.com, or CoP raw GitHub sources.GPL 3.0
- Identify a business from name, website, or short phrase. Returns business name, description, website, and country region to assess merchant before going live.MIT
- Queue a brand identity generation from a website URL. Notra scrapes the site and extracts brand information, allowing polling for completion.MIT
- Extract structured data from up to 10 URLs in a single call, storing rows directly into a ZeroDB table for efficient multi-page scraping.MIT