Connect your LLM to Just Publish.

Which endpoint

Three addresses, one server, one sign-in. Each address has its own OAuth token, because a token is bound to the endpoint that issued it. Pick by which assistant you are connecting, not by what you want to publish - the publishing behaviour is identical. A client that cannot sign in at all can use the bridge below.

Quick start

Add the endpoint your client needs, sign in, then ask the model to publish your files. Each tab below uses the endpoint from the table above. Want click-by-click connect steps instead of the raw schema? See Connect your AI assistant.

Server URL: https://mcp.justpublish.ai/claude
Authentication: OAuth 2.1 - Just Publish registers itself, so there is
                no client ID or client secret to fill in

{
  "mcpServers": {
    "just-publish": {
      "url": "https://mcp.justpublish.ai/"
    }
  }
}

Server URL: https://mcp.justpublish.ai/chatgpt
Authentication: OAuth

Once connected, ask the model something like: "Publish this HTML to Just Publish and give me the URL." The model calls deploy and hands back the live URL. The site belongs to your account, so the assistant can find it again with list_sites and edit it without a token.

The sign-in, in detail

https://mcp.justpublish.ai/, https://mcp.justpublish.ai/claude and https://mcp.justpublish.ai/chatgpt are one server behind OAuth 2.1, the same free sign-in as the website. A site published through any of them is filed under the signed-in person's account: deploy takes no email parameter, and the site is live immediately with nothing to prove afterwards. The discovery documents are https://mcp.justpublish.ai/.well-known/oauth-protected-resource, https://mcp.justpublish.ai/.well-known/oauth-protected-resource/claude and https://mcp.justpublish.ai/.well-known/oauth-protected-resource/chatgpt. They are separate endpoints so each gets its own audience-bound token, and each advertises the contract its directory's rules require: on /chatgpt, deploy creates a new site only and no tool takes an edit_token.

Because it knows who is asking, every endpoint serves list_sites, which takes no parameters and returns the websites that person has already published - live URL, site_id, whether each is live or archived, when it was last updated, and any connected custom domain. get_site_files and update_site_file work without an edit_token on that person's own sites.

Setting up the Claude connector, step by step and for every Claude surface: Connect Just Publish to Claude.

Clients that cannot sign in

Every endpoint needs a sign-in, so there is no open address. A client that cannot run an OAuth sign-in itself can use the mcp-remote bridge: it runs on your computer as a local MCP server, opens the browser for the sign-in once, and passes your requests to Just Publish.

{
  "mcpServers": {
    "just-publish": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://mcp.justpublish.ai/"]
    }
  }
}

Tool reference - deploy

Publishes a static site under the signed-in account. First call (no site_id) creates a new site. Later calls with the same site_id update that site in place. The schema below is the one served at https://mcp.justpublish.ai/ and /claude; on /chatgpt, deploy creates a new site only.

Parameters

File object

Response (success)

{
  "url":            "https://xK3p9wnA2b.justpublish.site/",
  "site_id":        "xK3p9wnA2b",
  "edit_token":     "...",            // only on create; optional for your own sites
  "created":        true,             // false on update
  "files_uploaded": 4,
  "total_bytes":    12842
}

Errors

Errors come back as a tool-call result with isError: true and a human-readable message - not as a JSON-RPC error. The model surfaces the message to the user.

JSON-RPC examples

Create a new site

POST https://mcp.justpublish.ai/
Accept: application/json, text/event-stream
Content-Type: application/json
Authorization: Bearer <access token from the sign-in>

{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "deploy",
    "arguments": {
      "files": [
        { "path": "index.html", "content": "<h1>hello world</h1>" }
      ]
    }
  }
}

Update an existing site

{
  "jsonrpc": "2.0",
  "id": 2,
  "method": "tools/call",
  "params": {
    "name": "deploy",
    "arguments": {
      "site_id": "xK3p9wnA2b",
      "files": [
        { "path": "index.html", "content": "<h1>hello v2</h1>" }
      ]
    }
  }
}

Constraints

• index.html at the root of the file map is required. A site without it is rejected at deploy.
• Max site size: 50 MB. Max single file: 5 MB. Max 500 files per site.
• Path rules: no .., no leading /, no backslash, no null bytes. The deploy is rejected at validation time, not silently truncated.
• Per-account rate limits on both site creation and deploys per rolling hour.
• Static files only. No server-side execution, no build step, no environment-variable injection.
• Sites with no traffic for an extended period may be archived (TTL policy).

Your sites & the edit token

Every site is filed under the account that signed in. list_sites returns them, and get_site_files, update_site_file and deploy work on them with just the site_id. The dashboard lists the same sites.

deploy still returns an edit_token when it creates a site. It is a second key to that one site, for a client that keeps it; the server stores only a hash, and you do not need it for your own sites.