{"id":"dev.urlpipe/mcp","slug":"dev-urlpipe-mcp","title":"URLpipe","description":"Read any page with its JavaScript run: Markdown, screenshots, metadata, console errors, Lighthouse.","version":"1.4.0","publisher":{"kind":"domain","id":"urlpipe.dev","label":"urlpipe.dev","url":"https://urlpipe.dev"},"website":"https://urlpipe.dev/mcp-server","repository":"https://github.com/URLpipe/mcp","remotes":[{"type":"streamable-http","url":"https://urlpipe.dev/mcp","headers":[{"description":"Bearer followed by a URLpipe organization token, created under Organization → MCP access at https://urlpipe.dev","isRequired":true,"format":"string","value":"Bearer {token}","isSecret":true,"variables":{"token":{"description":"A URLpipe organization token","isRequired":true,"isSecret":true}},"name":"Authorization"}]}],"packages":[],"published":"2026-10-12","updated":"2026-10-12","status":"online","auth":"","latency_ms":1873,"checked":"2026-11-19","uptime":100,"checks":3,"tools":[{"name":"list_projects","title":"List projects","description":"The projects this token can reach, each with how many requests it has\nmade in the last 30 days and where its results are delivered.\n\nStart here: every other tool takes a project_id from this list. A project\nis the unit a request is billed and recorded against — credits themselves\nbelong to the organization, so which project you pick does not change\nwhat a call costs.","read_only":true},{"name":"get_usage","title":"Check usage and prices","description":"What the organization's plan allows, how much of it is left this period,\nand what each operation costs.\n\nWorth reading before a run of expensive calls: the operations differ by\nseventeen times in price, so the difference between fetch_markdown and\nsummarize_page over a hundred pages is 100 credits against 1,700. It is\nalso how to read a refusal — a call that comes back with quota_exceeded\nis telli","read_only":true},{"name":"fetch_markdown","title":"Read a page as Markdown","description":"Exposes POST /markdown. Renders the page in headless Chrome and converts\nits main content to clean Markdown — headings, lists, links and code\nkept, navigation, sidebars and cookie banners dropped.\n\nStart here when you want to READ a page. It is the cheapest operation we\nsell (1 credit) and by far the most compact thing to put in front of a\nmodel: fetch_html returns the whole DOM, which is usually ","read_only":false},{"name":"fetch_html","title":"Get the rendered HTML","description":"Exposes POST /html. Returns the page's HTML after JavaScript has run and\nredirects have been followed — the DOM a real browser sees, not the empty\nshell curl returns. 1 credit.\n\nUse it when you need the markup itself: a specific attribute, a script\ntag, a structured-data block. If you want to read the page's content,\nfetch_markdown costs the same and returns a fraction of the text.","read_only":false},{"name":"capture_screenshot","title":"Take a screenshot","description":"Exposes POST /screenshot. Captures the whole rendered page as a PNG —\nJavaScript executed, web fonts and images included, exactly as a browser\nwould draw it. 1 credit, whichever options you use.\n\nUse it to see a page rather than read it: layout, visual regressions,\nlink previews, or checking what an anti-bot page actually showed us.\nscreenshot_options size it (viewport, scale, one element, the fol","read_only":false},{"name":"capture_pdf","title":"Save a page as PDF","description":"Exposes POST /pdf. Prints the whole rendered page to a PDF — JavaScript\nexecuted, web fonts and lazy-loaded images included, its headings as\nbookmarks. 1 credit, whichever options you use.\n\nUse it to archive or hand on a page as a document: a receipt, an article,\na report, terms as they stood today. pdf_options choose the paper, the\norientation, the margins and the scale, whether backgrounds print","read_only":false},{"name":"capture_console_errors","title":"Capture console errors","description":"Exposes POST /console. Loads the page in headless Chrome and returns\nwhat it reported through console.error and console.warn during load, plus\nuncaught exceptions and unhandled promise rejections (not console.log). 1 credit.\n\nWhat you would open DevTools for: broken third-party scripts and\nclient-side errors you cannot reproduce locally.","read_only":false},{"name":"run_lighthouse_audit","title":"Run a Lighthouse audit","description":"Exposes POST /lighthouse. Runs a real Google Lighthouse audit against the\nlive page: performance, accessibility, best-practices and SEO scores plus\nCore Web Vitals. 2 credits — dearer than a page fetch because the audit\nruns in its own limited lane, so it also takes longer than anything else\nhere.\n\nIt fetches the page itself rather than sharing a visit, so asking for it\ninside scrape_url costs two","read_only":false},{"name":"extract_metadata","title":"Extract page metadata","description":"Exposes POST /meta. Reads everything a page declares about itself —\nOpen Graph, Twitter cards, meta tags, JSON-LD, microdata and links — into\none object with the same keys on every page: title, description, site\nname, kind of page, language, authors, publication and update dates, share\nimage and video, favicon and icons, logo, feeds, keywords, canonical URL,\nrobots directives, hreflang alternates,","read_only":false},{"name":"detect_technologies","title":"Detect a page's technologies","description":"Exposes POST /technologies. Detects what a page is built with — CMS,\ne-commerce platform, JavaScript frameworks and libraries, analytics and\ntag managers, advertising and marketing tools, CDN, web server,\nprogramming language, payment providers, chat widgets, consent managers\nand more — from an open database of over 7,000 fingerprints. Each\ntechnology comes with its version when the page gives it ","read_only":false},{"name":"extract_keywords","title":"Extract keywords","description":"Exposes POST /keywords. Returns the 5–15 terms and phrases that best\nrepresent the page, ordered by relevance — ranked by a language model,\nnot by raw frequency. 15 credits.\n\nAs with summarize_page: if you are going to reason over the result\nyourself, fetch_markdown costs 1 credit and gives you everything.","read_only":false},{"name":"summarize_page","title":"Summarize a page","description":"Exposes POST /summarize. Returns a concise Markdown summary of the page's\nmain content, with the navigation, ads and boilerplate left out.\n\nThe most expensive operation we sell, at 17 credits, because it runs a\nlanguage model over the page. If YOU are the model that will read it,\nfetch_markdown gives you the whole page for 1 credit and you can\nsummarize it yourself — this tool is for when the summ","read_only":false},{"name":"extract_data","title":"Extract structured data with your own schema","description":"Exposes POST /extract. Reads the page and fills an object YOU define:\npass `fields`, each with a name, a type and a description of what to take\nfrom the page, and get back exactly those keys in that order. Anything the\npage does not give comes back null — values are never guessed.\nTypes: string, number, integer, boolean, date, datetime, url, email, object (with its own\nfields) and array (with item","read_only":false},{"name":"scrape_url","title":"Several results from one page visit","description":"Exposes POST /scrape. Runs any subset of the other operations in one\nrequest, served from a single page visit where possible. Each operation\nis billed and stored exactly as the individual call would be, so this is\nnot a discount — with one exception, and it is the reason to use it: a\nresidential exit is charged per page VISIT, so several results off one\nvisit pay the surcharge once.\n\nAnswers with ","read_only":false},{"name":"get_result","title":"Get a result","description":"Exposes GET /result/:token. Returns the result of a past request, in the\nsame form the call that made it would have returned — Markdown as\nMarkdown, a structured operation as its object, a screenshot as an image,\na PDF as an embedded application/pdf resource,\na scrape as its combined object.\n\nThis is how an async call is collected: any tool called without\nsync: true answers with a token, and this ","read_only":true},{"name":"list_requests","title":"List past requests","description":"A project's recent requests, newest first, with each one's token.\n\nLook here before fetching a page: if somebody already pulled it, passing\nthat token to get_result returns the same result for nothing. A /scrape\nappears as one entry — its per-operation children are internal, and the\nscrape's own token returns all of them.\n\nPass labels to see only the requests made with them — every key given\nmust ","read_only":true},{"name":"get_request","title":"Inspect a past request","description":"Everything about one past request except its result: what was asked for,\nhow long each stage took, whether it came from the store, and how the\nwebhook delivery went.\n\nThis is the metadata; get_result returns the page itself. Reach for this\none when a call did not do what you expected — it says whether the result\nwas reused, why an analysis failed, and whether the delivery to the\nproject's endpoint","read_only":true}],"server_info":{"name":"urlpipe","title":"URLpipe","version":"1.5.1"},"category":"browser","tags":["web-search","marketing","productivity"],"kind":"remote","source":"registry","submitted":{},"first_party":true,"domain_rank":null,"score":100,"indexable":true,"first_seen":"2026-09-25","stars":0,"badges":[],"sponsor":null,"page":"https://agentsindex.org/agents/dev-urlpipe-mcp/"}