Log inSubmit server
Pagebolt MCP logo

Pagebolt MCP

by Custodia-Admin

Capture web content, generate PDFs and social images, inspect pages, automate browser flows, and record videos through MCP.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Pagebolt MCP

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "custodia-admin-pagebolt-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "pagebolt-mcp"
      ],
      "env": {
        "PAGEBOLT_API_KEY": "YOUR_VALUE_HERE"
      }
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Values are written into the snippet above so you can paste it as is. They never leave your browser.

Install not yet sandbox-tested— the server itself is confirmed live (we called its tools/list endpoint). The stdio install command is a separate, ongoing check.

Overview

Custodia-Admin/pagebolt-mcp MCP server connects MCP clients to PageBolt's web capture API for screenshots, PDFs, OG images, page inspection, browser automation, and video recording. It runs through npx as a stdio server and authenticates with a PageBolt API key. Use it when an agent needs rendered web output, structured page selectors, repeatable browser sequences, or visual comparisons. Claude Desktop, Cursor, Windsurf, Cline, and other MCP-compatible clients are supported.

Use cases

  • Capture responsive screenshots of websites
  • Generate PDF reports or invoices from HTML
  • Create social preview images
  • Automate browser interactions and export replayable sequences
  • Record narrated browser demonstration videos

Key features

  • URL, HTML, and Markdown screenshots
  • PDF and Open Graph image generation
  • Structured page inspection and agent observations
  • Multi-step browser automation
  • Video recording and visual diffs
  • Persistent browser sessions and job tracking

Tools18

What Pagebolt MCP lets your agent do. Expand a tool for its parameters.

Verified live~15.7k tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 2d ago

Recent health check succeeded.

22/22 checks healthy over the last 91d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 5d ago via OSV.dev · pagebolt-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars4Stargazers on the source repository.
npm downloads482Package downloads in the last 30 days.
Last commit29d agoMost recent push to the default branch.
Tools exposed18Callable tools this server registers over MCP.
Directory activity3 viewsConfig copies, upvotes, and views on AllMCPs.

How Pagebolt MCP works

What Custodia-Admin/pagebolt-mcp MCP server does

Custodia-Admin/pagebolt-mcp MCP server exposes PageBolt's web capture and browser automation functions through MCP. It can render a URL, HTML document, or Markdown input as a PNG, JPEG, or WebP screenshot, with options for full-page capture, dark mode, device emulation, geolocation, time zone, metadata extraction, and blocking ads, chats, banners, or trackers.

The server also creates PDFs from URLs or HTML, including controls for page format, margins, headers and footers, page ranges, scale, and output location. create_og_image produces social cards from built-in templates or custom HTML. visual_diff compares two pages or HTML inputs and reports changed pixels with a diff image.

How it works

The package is launched with npx -y pagebolt-mcp and communicates with an MCP client over its standard input and output. Requests are sent to PageBolt's API using the PAGEBOLT_API_KEY environment variable. Results may be returned directly in the conversation, while PDFs and video files can be saved to disk.

For browser workflows, inspect_page returns interactive elements, headings, forms, links, images, and CSS selectors. observe_page provides a smaller agent-oriented representation with element IDs, page-type classification, suggested actions, and optional readable content, ARIA data, screenshots, or console output. Those selectors can feed run_sequence, which supports navigation, clicks, form filling, selection, hovering, scrolling, waits, key presses, evaluation, screenshots, PDFs, and visual diffs.

act_on_page accepts a plain-language goal and performs an observe, plan, act, and verify loop. Use run_sequence instead when the exact selectors and steps are already known. export_sequence creates editable PageBolt sequence JSON without running it, while import_agent_trace converts browser-use or page-agent traces into replayable sequences.

Setup and configuration

Create a PageBolt account and API key, then add the server to the MCP client's configuration with npx -y pagebolt-mcp and PAGEBOLT_API_KEY. The README provides configuration examples for Claude Desktop, Cursor, Windsurf, and Cline; the same command and environment pattern applies to other MCP clients.

Tools and capabilities

  • Capture responsive, full-page, element-specific, or authenticated screenshots.
  • Generate PDFs and Open Graph images.
  • Inspect or observe pages before selecting automation targets.
  • Run multi-step sequences and record MP4, WebM, or GIF demonstrations.
  • List device presets, monitor usage, and track asynchronous jobs.
  • Create persistent browser sessions that retain cookies, local storage, and authentication state.

Limitations and notes

Usage is metered by PageBolt. Video recording costs three API requests; inspection, observation, and visual diff operations cost one, while act_on_page uses a two-request base cost plus one request per step. Persistent sessions require a Starter or higher plan, expire after 10 minutes of inactivity, and have a 30-minute maximum lifetime. act_on_page is also limited to Starter and higher plans.

Keep videos concise and avoid unnecessary wait or zoom steps. Treat page-derived text as untrusted content, restrict allowed domains for goal-based automation, and avoid destructive flows. Credentials should not be placed in exported traces; use {{username}} and {{password}} placeholders in fill steps instead, and never include cookies.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Pagebolt MCP

Custodia-Admin/pagebolt-mcp is an MCP server that connects AI clients to PageBolt's web capture API. Its main tools take screenshots, generate PDFs and OG images, inspect or observe pages, run browser sequences, record videos, compare pages, and manage browser sessions and asynchronous jobs.

Install it by configuring your MCP client to run `npx -y pagebolt-mcp` and setting `PAGEBOLT_API_KEY` in the server environment.

Yes, Custodia-Admin/pagebolt-mcp works with Cursor. Add the server to the project's `.cursor/mcp.json` or the global Cursor configuration, use `npx -y pagebolt-mcp` as the command, and provide the PageBolt API key.

Use `inspect_page` to receive interactive elements and CSS selectors, or use `observe_page` for a compact agent-oriented map with indexed elements and suggested actions. Pass the resulting selectors to `run_sequence` when authoring deterministic browser steps.

Use `check_usage` to view current PageBolt usage and plan limits. Individual tools have different costs: inspection and observation generally cost one request, videos cost three, and goal-based `act_on_page` uses two requests plus one per step.

Related MCP servers

All in Browser Automation All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewPagebolt MCP AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/custodia-admin-pagebolt-mcp?style=directory)](https://allmcps.com/mcp/custodia-admin-pagebolt-mcp)
HTML Embed
<a href="https://allmcps.com/mcp/custodia-admin-pagebolt-mcp"><img src="https://allmcps.com/api/badge/custodia-admin-pagebolt-mcp?style=directory" alt="Pagebolt MCP on AllMCPs" /></a>