Browser Use
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
제작자가 쓴 설명(영어 원문)
Create, edit, preview, publish, and manage web pages from MCP-capable AI clients.
이 세 가지는 절대 하나의 점수로 합치지 않아요. 의견, 측정, 인기는 서로 다른 것이니까요.
이 서버는 아직 쉬운 말로 된 요약을 쓰지 못했어요. 아래는 레지스트리에서 읽어 온 사실이에요.
쓰고 있는 AI 앱을 고른 다음 설정을 복사하세요.
필요한 것
어떤 앱을 쓰세요?
위에서 앱을 고르세요. 기억해 두었다가 다음부터는 버튼에 “…에 설치”라고 표시해요.
웹, 데스크톱, 모바일의 Claude에서 쓸 수 있어요:
https://create-web-page.com/mcp터미널에서 이 명령어를 실행하세요
claude mcp add --transport http create-web-page https://create-web-page.com/mcp프로젝트의 .cursor/mcp.json 또는 ~/.cursor/mcp.json에 붙여 넣으세요
{
"mcpServers": {
"create-web-page": {
"url": "https://create-web-page.com/mcp"
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
프로젝트의 .vscode/mcp.json에 붙여 넣으세요
{
"servers": {
"create-web-page": {
"type": "http",
"url": "https://create-web-page.com/mcp"
}
}
}브라우저가 앱을 열지 물어본 다음, 앱이 설정을 보여 주고 확인을 요청해요. 키 값은 비워 두니 직접 입력하세요.
ChatGPT는 인터넷에 있는 서버에만 연결할 수 있어요:
https://create-web-page.com/mcp메뉴 이름은 요금제나 시기에 따라 조금 다를 수 있어요.
선택한 앱은 이 기기의 쿠키에만 저장돼요. 계정은 필요 없어요.
이 설정은 제작자의 레지스트리 정보로 만들었어요. README와 다르면 README를 따르세요.
도구 이름과 설명은 서버가 2026년 11월 10일에 알려 준 그대로예요. 우리가 아니라 제작자가 쓴 글이고, 일반 텍스트로만 보여 드려요.
account_whoamiReturn the current Create Web Page account context. Use this to check whether the user is authenticated or using an anonymous expiring demo.
page_intake_startOptional first step for a vague page request (e.g. 'make me a website for my barbershop'). Returns at most one round of questions to ask together: the name, the main visitor action AND where it goes (WhatsApp number with country code, booking/order link, phone or address), and mu
page_onboarding_startGuided visual setup (category, layout, palette, primary action) that does NOT create a page. Use it ONLY when the user explicitly wants to choose the style/layout step by step; for normal requests call page_create_from_brief (asking at most one round of questions first). IMPORTAN
page_onboarding_updateValidate onboarding selections and return the remaining checklist. STATELESS: it does not remember previous calls, so every call must include the FULL accumulated selection (mainCategory, subCategory, layout, palette, primaryAction) plus all knownFields (name, description, action
source_importFetch a public website URL and extract a structured brief with business name, description, contact links, and suggested page blocks. Only works with public HTTP/HTTPS pages. Does not bypass login, CAPTCHA, or paywalls. Returns extracted fields and suggestedBlocks for use with pag
page_createLow-level/legacy: create a blank link-in-bio scaffold from a style preset (title, displayName, preset or 'auto'). Not for normal page requests — use page_create_from_brief, which builds the page from the user's details. The scaffold starts with empty placeholder blocks to fill vi
page_create_from_briefThe main tool for creating a page. Call it directly when the request already has the business/page name and the main visitor action with its real destination (WhatsApp number with country code, booking/order link, phone, address or email); if the request is vague, ask at most one
block_addAppend a new content block to a link-in-bio page. The block must match one of the 14 allowed types (link, social_row, whatsapp, phone_call, email_contact, image, gallery, video_embed, text, map_embed, product_card, lead_form, reviews, save_contact) with that type's strict props.
layouts_listList supported page layouts. Currently ships link_in_bio only.
presets_listList the 12 curated style presets (theme + starter blocks) for link-in-bio. Use to offer the user a choice or to confirm a preset:auto selection.
page_getRead a landing page draft, including its latest content JSON. For anonymous demo pages, include the editToken.
media_uploadAdd an image to a page. PREFERRED for any image already on the web: pass its imageUrl and the server fetches it directly (no size limit, no attachment needed in chat). imageBase64 is ONLY for very small inline images — large base64 payloads are rejected by request-size limits and
media_upload_link_createGet an upload link for a photo on the user's own device (phone/computer) — the right tool whenever the image is not already at a public URL. Pass pageId (+ editToken for demo pages) and the placement (e.g. header_photo). Then show the returned uploadPageUrl to the user as a click
media_attachAttach a ready media asset to the page. placement determines where: avatar (header), block_image (image inside a specific block — requires blockId), gallery_item (appended to a gallery — optional blockId targets a specific gallery; otherwise the first one is used), background (th
media_deleteSoft-delete a media asset and remove all references to it from the page content.
page_claimClaim an anonymous demo page into an AUTHENTICATED account. This requires the signed-in user's OAuth — an anonymous chat/agent session cannot call it and will get an OAuth-required error. So in a normal agent conversation, do NOT call this to 'save' a demo page: instead share the
account_pages_listList landing pages saved to the signed-in account. Safe to call without sign-in: an anonymous chat gets ok=true, authenticated=false and an empty list (not an error) because it has no saved pages — then create the page directly with page_create_from_brief. Sign-in is only needed
account_page_manageRead a claimed account page without requiring an editToken. Use after OAuth when the user wants to manage a durable website.
page_header_updateUpdate the link-in-bio page header: displayName, bio, verified checkmark, business-card display style ("photo_hero" large top photo or "avatar"), a short role label (e.g. "Barber · Madrid"), labelPosition, and the header photo (photoUrl; null removes it). Use only the fields you
page_theme_updateApply a partial theme patch. Theme fields are enum-constrained (button variant/radius/shadow, font family/weight, density, icon_size). Use presets_list first if the user wants a complete restyle. For already-published pages, this saves an unpublished latest revision only. Do not
page_meta_updateUpdate internal page title, URL slug, and SEO title/description. For already-published pages, this saves an unpublished latest revision only. Do not call page_publish in the same assistant turn after this edit. Stop and tell the user the draft/preview was updated, then wait for a
page_previewReturn the current preview URL for a page. You usually do not need this: page_create and page_create_from_brief already return a ready-to-open previewUrl. For an anonymous demo page you must pass the editToken from the create result; calling this with pageId alone fails.
page_publishPublish the latest page revision. Call this only when the user's latest message explicitly asks to publish, make the page live, or get a public share link. Do not call this immediately after page_create or page_create_from_brief just because the draft is publish-ready. For anonym
page_unpublishRemove a published landing page from public availability while keeping its draft and revision history.
block_updatePatch one content block by stable blockId. Provide a props patch (merged into existing props) and/or a new type. Block types and props are validated against the strict per-type schema. For already-published pages, this saves an unpublished latest revision only. Do not call page_p
block_deleteDelete one block by stable blockId. Use page_get first if unsure which id to remove. For already-published pages, this saves an unpublished latest revision only. Do not call page_publish in the same assistant turn after this edit. Stop and tell the user the draft/preview was upda
block_reorderMove a block to a new zero-based index in the page main slot. For already-published pages, this saves an unpublished latest revision only. Do not call page_publish in the same assistant turn after this edit. Stop and tell the user the draft/preview was updated, then wait for a se
누구나 볼 수 있는 정보를 점검했고, 코드는 읽지 않았어요. 해당하지 않는 항목은 개수에서 뺐어요. · 2026년 11월 10일 점검
게시자(create-web-page.com)가 저장소 소유자(daniel-lopez-puig)가 아니에요. 다른 사람이 등록했을 수 있어요
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
아직 GitHub에서 이 저장소를 읽어 오지 않았어요
90ms 만에 응답(HTTP 200)
로그인 필요 없음
도구 설명 27개를 읽었고, AI에게 보내는 숨겨진 지시는 없었어요
레지스트리에서 이 카테고리의 활성 서버 전체와 비교했어요. 스타는 게시자가 저장소를 소유한 경우에만 세요.
해당하는 기본 점검 3개 중 2개를 통과했어요. 코드는 읽지 않았어요. 키나 토큰을 넣기 전에 서버가 어떤 권한을 요구하는지 확인하고, 꼭 필요한 권한만 있는 키를 쓰세요.
레지스트리에 필수 키가 나와 있지 않아요. 다만 처음 쓸 때 로그인을 요구하는 서버도 있어요.
제작자의 클라우드에서 실행돼요. 설치 없이 인터넷으로 연결하지만, 보내는 내용은 제작자의 서버를 거쳐요.
GitHub 저장소는 마지막으로 – 변경됐고, 최신 버전(0.1.0)은 6개월 전 레지스트리에 올라왔어요.
Claude Desktop, Claude Code, Cursor, VS Code처럼 MCP를 지원하는 앱에서 쓸 수 있어요. URL로 연결되기 때문에, 원격 MCP 서버를 지원하는 앱(예: 개발자 모드의 ChatGPT)에서도 쓸 수 있어요.
같은 카테고리에는: Browser Use, Scrapling MCP Server, Chrome DevTools MCP
같은 카테고리에서 다른 제작자가 만든 서버예요. 스타 순으로 정렬했어요.
카드 보는 법:↓ 주간 다운로드i★ GitHub 스타i세 기둥: 리뷰 · 점검 · 사용 정도i업데이트 = 최신 버전을 레지스트리에 공개한 날i
Browser agents and stealth cloud browsers for AI: run web tasks, fill forms, extract data.
Web scraping with stealth HTTP, real browsers, and Cloudflare bypass. CSS selectors supported.
저장소에서 README를 찾지 못했어요