mcpField
P
PageProofer

PageProofer

Remote
pageproofer.comDeveloper Tools0Updated 3w ago

About PageProofer

PageProofer's hosted MCP server lets coding agents work with visual website feedback notes. Agents can inspect screenshots and browser errors, make code fixes, comment on changes, and send notes for human review.

Key features

  1. 01List sites and filter feedback notes
  2. 02Read note context and browser errors
  3. 03View screenshots and attachments
  4. 04Comment on notes and update fields
  5. 05Move fixed notes to in review

Use cases

  1. 01Triage open website feedback
  2. 02Inspect a screenshot and console errors
  3. 03Fix a reported page element issue
  4. 04Document changes and request review

MCP support

The hosted remote server connects to MCP clients over Streamable HTTP and requires a bearer token. It exposes tools for listing sites and notes, retrieving attachments, commenting, and updating note details and status.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://app.pageproofer.com/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://app.pageproofer.com/mcpAuth required

Tools

Details

From the project README

PageProofer MCP server

Read, fix and reply to PageProofer website feedback notes from GitHub Copilot, Claude Code, Cursor and other AI agents.

PageProofer collects visual feedback on websites: clients and teammates click an element, leave a note, and PageProofer captures the page, the element, the browser and any console errors. This MCP server lets your coding agent work through those notes. It finds the issue in your code, fixes it, comments on the note to say what changed, and moves the note to review.

This is a hosted (remote) server, so there's nothing to install:

  • URL: https://app.pageproofer.com/mcp
  • Transport: streamable HTTP
  • Auth: Authorization: Bearer YOUR_TOKEN

This repository only documents the server. The server itself runs as part of PageProofer.

What the agent can do

ToolWhat it does
list-sitesList the websites you can work on, with their domains, open note counts and the people on each site.
list-notesList notes on a site, newest first. Filter by status, #tags, text, page, assignee or overdue. Open notes only by default.
get-noteEverything about one note: the feedback, page URL, the element it was pinned to (CSS selector and HTML), browser and window size, console errors, failed network requests, comments and attachments.
get-attachmentFetch a screenshot or other attachment. Images come back as images the agent can see, text files as their contents.
add-commentComment on a note as you, for example to say what was fixed and in which commit. Can be internal (team only).
update-noteChange a note's status, priority, assignee or due date.

Agents can't mark notes complete or closed. When an agent has fixed something, it moves the note to in review so a person can check it.

Get a token

  1. Sign in to PageProofer, open My Settings and go to the AI agents tab.
  2. Name the token (for example "VS Code on my laptop") and click Create token.
  3. Copy it straight away. It's only shown once.

AI agent access is included on every PageProofer plan, including the free trial. You can create a token per computer or tool and revoke any of them at any time.

Configure

GitHub Copilot in VS Code

Add this to .vscode/mcp.json in your project, or to your user MCP configuration. VS Code asks for the token the first time and stores it securely.

{
  "inputs": [
    {
      "type": "promptString",
      "id": "pageproofer-token",
      "description": "PageProofer token (My Settings > AI agents)",
      "password": true
    }
  ],
  "servers": {
    "pageproofer": {
      "type": "http",
      "url": "https://app.pageproofer.com/mcp",
      "headers": {
        "Authorization": "Bearer ${input:pageproofer-token}"
      }
    }
  }
}

Then open Copilot Chat in Agent mode and ask it about your notes.

Claude Code
claude mcp add --transport http pageproofer https://app.pageproofer.com/mcp --header "Authorization: Bearer YOUR_TOKEN"
Cursor

In ~/.cursor/mcp.json or .cursor/mcp.json:

{
  "mcpServers": {
    "pageproofer": {
      "url": "https://app.pageproofer.com/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_TOKEN"
      }
    }
  }
}
Other agents

Any MCP client that supports remote servers over HTTP can connect with the URL https://app.pageproofer.com/mcp and the header Authorization: Bearer YOUR_TOKEN.

Use it

Ask in plain language, from your project's folder:

  • "What are the open notes on the Northwind site?"
  • "Fix the open #checkout notes, then comment on each with what you changed."
  • "What's wrong with note 12? Look at the screenshot."
  • "Show me overdue notes assigned to me."

A typical loop is: the agent lists the open notes, reads one with its screenshot and console errors, finds the code behind the element, makes the fix, comments with the commit, and moves the note to in review.

How it acts

  • The agent acts as you. It sees the sites and notes you can see, nothing more.
  • Its comments and changes show your name, marked "via AI agent".
  • Notifications work as usual, so a comment from the agent notifies the people on the note.
  • Note text and comments come from clients and site visitors. The server tells the agent to treat them as descriptions of a problem, not as instructions.

Treat your token like a password. Don't commit it to a repository. If it might have leaked, revoke it in My Settings > AI agents.

Links

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.