Jaybird Digital
● Activecom.jaybirddigital/website
Build and publish a website from a chat. No account or API key needed; you get a live link to share.
Connect this server
One click to install, or copy the config below. Suits: Cursor · HTTP clients.
Or add it manually — claude_desktop_config.json (Claude Desktop) or .cursor/mcp.json (Cursor):
{
"mcpServers": {
"website": {
"url": "https://jaybirddigital.com/api/start/mcp"
}
}
}claude mcp add --transport http website https://jaybirddigital.com/api/start/mcpcode --add-mcp '{"name":"website","type":"http","url":"https://jaybirddigital.com/api/start/mcp"}'Syntax from each client's docs (Claude Code: claude mcp add; Codex: codex mcp add; VS Code: code --add-mcp). Placeholders in angle brackets are values you fill in. Codex takes hosted servers in config.toml ([mcp_servers.<name>] url = …) rather than on the command line.
A starting point — check the repository for exact arguments and setup. Restart your client after editing the config.
- Category
- Search & Web
- Version
- 1.3.3
- Registry
- active
- Published
- 11/21/2026
Is it safe? Is it maintained?
Serverdex check: Nov 22, 2026Looks healthy: Jaybird Digital is maintained, open source, hosted endpoint answering without a key.
- Maintained
- Active
- Last commit today (Nov 21, 2026), 0 open issues.
- Official registry
- Active
- Published as com.jaybirddigital/website in the official MCP registry, version 1.3.3.
- Source
- Open source, MIT
- Public repository under the MIT licence, 0 stars, 0 forks.
- Runs
- Hosted by the publisher
- Your client talks to jaybirddigital.com: prompts and tool arguments leave your machine for that host.
- Credentials
- None needed to connect
- The hosted endpoint completed the Serverdex initialize + tools/list handshake with no key (checked Nov 22, 2026); individual tools may still ask for one.
- Owner
- Unclaimed
- Nobody has claimed this page yet; the data above comes from the registry, GitHub and the live check.
Facts only: last commit and licence from GitHub, the entry's status from the official registry, credentials from the live check. Not a code audit — run the security scan on the endpoint yourself and see the status history.
Live check
checked Nov 22, 2026, 06:07 UTCStatus history & uptime →The endpoint answered the MCP handshake as "Jaybird Digital" and listed 12 tools. Answering since Nov 22, 2026.
Context cost ≈ 5.3k tokens of tool definitions a client loads per conversation (estimate: characters ÷ 4) across 12 tools.
Tools (12)
start_website2 paramsCreates a new, empty website and returns its websiteId, the editKey for changing it, the viewUrl where it can be seen, the keepUrl where the person keeps it, photoLink (a link where the person adds their own photos), siteZip (an address an AI that can run commands sends the whole website to as one zip, photos included), and the rules and size limits its files must meet. Each call makes a separate website; to change one already started, save to its websiteId instead. No account is needed: a draft lasts 7 days unless the person keeps it, and with an account connected it joins that account.
Show parametersHide parameters
businessNamestringrequiredThe website's name as the person gave it, such as their business's name. It is shown in the website's address.aboutstringOptional: what the person said the website is for, in their words.
save_website_files3 paramsAdds or replaces files on a website and checks every page against the rules start_website returned. Files not sent stay as they are; delete_website_file removes one, and edit_website_file changes part of one. Never send a photo here as base64: one photo is hundreds of thousands of characters, slow and costly to write; the person adds photos at the photoLink start_website returned, and save_website_pictures saves pictures you hold as files or at an https address. If you can run commands with internet access, write the website in a folder and send it all at once, photos included, to siteZip instead of typing every file here. Returns the viewUrl, keepUrl, pages, passed, and problems naming each rule a page breaks; the person can keep the website once passed is true. A website holds up to 40 files and 12 MB, with pages up to 300 KB and other files up to 2 MB.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.filesarrayrequired1 to 40 files to add or replace. A file whose path is already saved is overwritten.
edit_website_file4 paramsChanges part of one saved page, stylesheet or script in place: each edit replaces its find text with its replace text, so a fix sends only the text it changes instead of the whole file. Copy find exactly from the file (read_website_file shows it); it must appear once, or set all to change every one. Edits apply in order, and if any fails nothing is saved. Every page is then checked against the rules, and it returns the same report as save_website_files with edited. Fails with not_in_file when a find text is not there, found_more_than_once when it is there several times, and not_text for a picture or font. To add a file or replace a whole one, use save_website_files.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.pathstringrequiredThe file's path from the website's root, exactly as it was saved, such as index.html, about.html or images/shop.jpg. list_website_files shows every path.editsarrayrequired1 to 20 changes to the file, applied in order.
save_website_pictures4 paramsSaves pictures the person gave you, such as photos they uploaded in the conversation, as files on a website, without sending their bytes as text: pass each one as a file in pictures and it is downloaded from its download_url. ChatGPT fills the file in itself; any AI can also pass a picture's public https address as download_url, with file_id any name for it. Each is saved as images/<its file name> unless paths names another place, and a large picture is made smaller to suit a web page. Returns saved, each picture's path and size to use as an <img> src in the pages, beside the same report as save_website_files. Up to 10 pictures a call. Fails with picture_unreachable when a download address has expired (pass the file again) and not_a_picture for anything but a JPEG, PNG, WebP or GIF. For a website with a plan, use upload_image instead.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.picturesarrayrequired1 to 10 pictures, each a file: ChatGPT fills in download_url and file_id for a file the person uploaded; another AI passes a picture's public https address as dpathsarrayOptional: where to save each picture, in the same order as pictures, such as images/gallery-1.jpg. The extension follows the picture's real type.
get_picture_upload_link2 paramsMakes a link where the person adds their own photos to a website from their phone or computer, so the photos never pass through the conversation. Give them uploadUrl: it takes up to 10 photos of up to 20 MB each and works for 24 hours. Each photo is saved as images/<its file name>, made smaller to suit a web page. When the person says they are done, list_website_files shows the saved paths to use as the src of an <img> in the pages. Making the link changes nothing on the website. For photos already passed to you as files, save_website_pictures saves them directly.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.
list_website_files2 paramsLists every file saved on a website, with each path and its size in bytes, beside the viewUrl, keepUrl, pages, passed and any problems. It changes nothing. The paths are the ones read_website_file and delete_website_file take; read_website_file on a photo shows it to you with its width and height. It also returns siteZip, an address that gives the whole website as one zip (curl -o site.zip) and takes one back (curl -T site.zip), for an AI that can run commands.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.
read_website_file3 paramsReturns one saved file from a website with its path, encoding and content: pages, stylesheets and scripts as utf8 text, and fonts as base64. A picture (such as a photo the person added) comes back as encoding picture with its contentType, bytes, width and height and no content; a smaller copy is attached as an image for you to look at, so you can see what it shows. It changes nothing. A path that is not saved fails with no_such_file; list_website_files shows which paths exist.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.pathstringrequiredThe file's path from the website's root, exactly as it was saved, such as index.html, about.html or images/shop.jpg. list_website_files shows every path.
check_website_on_phone4 paramsOpens a website's pages in a real browser at a phone's size (390 by 844, as an iPhone) and says, page by page, what a phone visitor would hit: sideways scrolling and what is too wide, pictures that do not load, text under 12px, links and buttons under 24px to tap, a missing or zoom-blocking viewport line, bars that cover the screen, and a main heading pushed below the first screen. Each page gets fine and findings in plain words. It changes nothing. Checks every page up to 5 (or the pages named), a few seconds each, up to 40 pages a day per website. With picture true it also returns a screenshot of the first page checked, for an app that can look at images. Fix what it finds with edit_website_file or save_website_files, then check those pages again. Fails with check_limit when the day's checks are used.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.pagesarrayOptional: up to 5 saved pages to check, such as ["index.html", "menu.html"]. Leave it out to check every page, home page first, up to 5.picturebooleantrue to also get a screenshot of the first page checked, as a phone shows it. Leave it out if your app cannot look at images.
search_domains1 paramAsks a domain registrar, live, which names can be registered, and returns up to about a dozen, each with its name, whether it is available and whether it is premium. Availability can change at any moment. Nothing is bought or held; a name is bought from the website's keep page.
Show parametersHide parameters
querystringrequiredA whole domain name such as sweetcrumbs.com, or a word to build names from such as sweetcrumbs.
delete_website2 paramsPermanently deletes a whole website: every file, its address and its keep link. It cannot be undone. Call it only when the person has asked, in their own words, for the website to be deleted; never to start over, since save_website_files and edit_website_file change a website in place. To remove one file instead, use delete_website_file. A website with a plan is refused with has_a_plan; the person removes it from their dashboard.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.
delete_website_file3 paramsRemoves one saved file from a website, then checks the remaining pages against the rules start_website returned, and returns the same report as save_website_files. Other files stay. A path that is not saved fails with no_such_file. Removing index.html leaves the website with no home page until one is saved. To delete the whole website, use delete_website.
Show parametersHide parameters
websiteIdstringrequiredThe websiteId start_website returned, or one list_sites shows.editKeystringThe editKey start_website returned, which proves the website is the caller's. Leave it out for a website in the connected account.pathstringrequiredThe file's path from the website's root, exactly as it was saved, such as index.html, about.html or images/shop.jpg. list_website_files shows every path.
list_sitesLists the websites in the person's Jaybird Digital account, sorted by name, with each one's id, name and live address. Websites started with start_website while the account was connected, or opened with their editKey, are listed under startedWebsites (up to 20, kept ones first). An account with none gets an empty list. Needs the person to connect their account; called without one, it asks them to sign in.
Checked by Serverdex with a plain initialize + tools/list call to jaybirddigital.com/api/start/mcp, no credentials, about once a day. A failed check is never a removal.
On GitHub
read Nov 22, 2026- Stars
- 0
- Forks
- 0
- Last push
- Nov 21, 2026
- License
- MIT
agent-skillsai-agentsai-website-builderclaude-code-pluginclaude-skillsgemini-cli-extensionlanding-pagellms-txtmcpmcp-servermodel-context-protocolno-codeopenapiremote-mcpskillssmall-businessstreamable-httpweb-hostingwebsitewebsite-builder
From the README
Jaybird Digital MCP server Connect your AI tool to Jaybird Digital, and it can build a website for you and hand back a live link. No account, sign-in, API key or payment is needed to start. It works in Claude, ChatGPT and other AI tools that let you add a connector or an MCP server. Add this address: Jaybird Digital builds, hosts and manages websites for small businesses. It is based in San Antonio, Texas. The server is hosted by Jaybird Digital. This repository holds setup steps, config files and an agent skill, not the server's code. • Add it to your AI tool • What your AI tool can do • What happens to the draft • Try it • For developers • Agent skill • Privacy and terms • Support Add it to your AI tool Add the address above in your AI tool's connector or MCP settings. You do not need to sign in. If you already have a Jaybird Digital account, see If you have an account. Ready-made config files for each tool are in examples/. Claude (web, desktop and mobile) 1. Open Customize > Connectors, choose Add, then Add custom connector. 2. For the name, enter Jaybird Digital. For the URL, enter https://jaybirddigital.com/api/start/mcp. Choose Continue. 3. If Claude shows the settings it detected, choose Continue again. 4. Under Authentication, choose Sign in when needed. You can use it without an account, and sign in later if you get one. Under OAuth client, keep the option marked Detected, or choose Register automatically. 5. Choose Add.…
Read the full README on GitHub →Remotes
- streamable-http
https://jaybirddigital.com/api/start/mcp
Alternatives to Jaybird Digital
All Jaybird Digital alternatives →Best Search & Web servers →The most-starred Search & Web servers in the index, with their live-check result.
| Server | Stars | Last push | Live check |
|---|---|---|---|
| Browser Use | 121,623 | Nov 2026 | Working |
| Scrapling MCP Server | 104,822 | Nov 2026 | Local / not checked |
| Chrome DevTools MCP | 58,028 | Nov 2026 | Local / not checked |
| Chrome DevTools MCP | 58,014 | Nov 2026 | Local / not checked |
| com.puter MCP Server | 44,182 | Nov 2026 | Needs sign-in |
Jaybird Digital: questions people ask
- Is Jaybird Digital a remote or a local MCP server?
- According to its entry in the official MCP registry, Jaybird Digital ships as a hosted endpoint (streamable-http) that your client connects to by URL.
- Does Jaybird Digital need an API key?
- Its hosted endpoint completed the Serverdex live check — an MCP initialize + tools/list call — with no credentials on Nov 22, 2026. Individual tools may still ask for a key when called.
- Is Jaybird Digital safe to use?
- Looks healthy: Jaybird Digital is maintained, open source, hosted endpoint answering without a key. In detail — maintained: Active; official registry: Active; source: Open source, MIT; runs: Hosted by the publisher; credentials: None needed to connect. Serverdex check of Nov 22, 2026. This is a reading of the repository, the registry entry and the live check, not a code audit; run the Serverdex security scan on the endpoint before trusting it with real data.
- Is Jaybird Digital still maintained?
- Active: the repository's last commit today. Last push Nov 21, 2026, 0 GitHub stars, MIT licence. Read from GitHub on Nov 22, 2026.
- What tools does Jaybird Digital expose?
- 12 tools as listed by the server itself during the live check, including start_website, save_website_files, edit_website_file, save_website_pictures, get_picture_upload_link. The full list with parameters is in the Live check section above.
- How do I add Jaybird Digital to Claude, Cursor or VS Code?
- Copy the config in "Connect this server" into your client's MCP config file (claude_desktop_config.json, ~/.cursor/mcp.json or .vscode/mcp.json), or use the one-click Add to Cursor / Add to VS Code buttons and the one-line commands for Claude Code and Codex; then restart the client.
More Search & Web servers
Browse all Search & Web →LaneRouter
lanerouter
Live road conditions for the agent web: check a site or API before acting, report after.
Laraue Boards
Laraue
List, view, create, edit, delete, link, comment on, and see the history of Laraue Boards issues.
Lazynext Accessibility Checker
Lazynext-AI
Scan URLs or HTML for WCAG 2.2 violations. 75-rule manifest, weighted score, shareable reports.
Lazynext AI Slop Checker
lazynextai
Detect AI slop in text, HTML, pages, and project folders. Free, no API key.
Linkbreakers MCP
linkbreakers-com
QR codes and links that capture, qualify and route leads: create codes, read who scanned, via MCP.
Linkjar
rocketgems
Send PDFs, decks and HTML pages as branded, tracked links, and see who opened them.
Add the badge to your site
Show visitors that Jaybird Digital is listed here — it’s also the link we check for a free listing.
HTML — for your website
<a href="https://serverdex.io/server/com-jaybirddigital-website?utm_source=badge"><img src="https://serverdex.io/brand/badge.svg" alt="Jaybird Digital — listed on Serverdex" width="123" height="40" loading="lazy"></a>Recommended. A dofollow link from your own domain is the one that counts for search engines, and it's the link we check for a free listing.
Markdown — for your README
[](https://serverdex.io/server/com-jaybirddigital-website?utm_source=badge)Nice for visibility, but GitHub marks README links nofollow, so it doesn't pass authority. Use the HTML badge on your site as well.
Install buttons — for your README
[](https://serverdex.io/install/cursor/com-jaybirddigital-website) [](https://serverdex.io/install/vscode/com-jaybirddigital-website) [](https://serverdex.io/install/claude-code/com-jaybirddigital-website)One click opens Cursor or VS Code with this server's config prefilled; the Claude Code button lands on the one-line command. The links point at Serverdex, so when your config changes the buttons keep working.
Install buttons — HTML for your docs
<a href="https://serverdex.io/install/cursor/com-jaybirddigital-website"><img src="https://serverdex.io/brand/install-cursor.svg" alt="Add to Cursor — via Serverdex" width="140" height="28"></a>
<a href="https://serverdex.io/install/vscode/com-jaybirddigital-website"><img src="https://serverdex.io/brand/install-vscode.svg" alt="Add to VS Code — via Serverdex" width="147" height="28"></a>
<a href="https://serverdex.io/install/claude-code/com-jaybirddigital-website"><img src="https://serverdex.io/brand/install-claude-code.svg" alt="Open in Claude Code — via Serverdex" width="180" height="28"></a>Same buttons as plain links for a docs page or website.
Prefer a live badge with the score, stars and live check? Score badge (links to a page that explains the number and back to this listing).