Read a URL as clean markdown, screenshot a website, url to PDF. Web access for agents, no signup.
mcp-name: dev.pagelens/pagewatch
An MCP server that gives an agent a real browser. Three tools:
read_page turns any url into clean markdownscreenshot returns a png of the rendered pagepdf turns a url or raw html into a pdfPlus register and balance for managing your key.
It runs javascript, so it works on pages that a plain http fetch returns empty. It respects robots.txt and refuses sites that block automation instead of trying to defeat them.
Your first call mints a free trial key and returns real content. There is no form for your agent to get stuck on. When the trial runs out, one email confirmation by a human unlocks 200 more free credits, still free.
Billing is metered, but there is no payment processor connected right now, so nothing is ever charged.
Add this to your MCP client config:
{
"mcpServers": {
"pagewatch": {
"command": "uvx",
"args": ["pagewatch-mcp"]
}
}
}
That is it. uv fetches the package on first run, and your first tool call returns real content.
Or install it yourself with pip:
pip install pagewatch-mcp
{
"mcpServers": {
"pagewatch": { "command": "pagewatch-mcp" }
}
}
If you would rather not run anything, the same tools are live at
https://pagelens.dev/mcp as a remote MCP server.
Everything is optional.
| Variable | Default | What it does |
|---|---|---|
PAGEWATCH_API_KEY | none | An existing key. Without one, the first call mints a free trial key and hands it back. |
PAGEWATCH_BASE_URL | https://pagelens.dev | Point at a different pagewatch deployment. |
PAGEWATCH_MCP_TRANSPORT | stdio | Set to http to serve streamable-http on $PORT instead. |
PAGEWATCH_SOURCE | mcp-selfhost | Attribution tag recorded when a key is registered. |
It is deliberately thin. Every tool is one https call to a documented /v1 route
on the pagewatch API. There is no browser, no scraping logic and no credential
store in this package. You can read the whole thing in a couple of minutes and
know exactly what it sends and where.
docker build -t pagewatch-mcp .
docker run --rm -i pagewatch-mcp
MIT
Source-derived launch command. Check the maintainer’s required arguments and credentials before running:
npx -y pagewatch-mcpMerge this template into ~/Library/Application Support/Claude/claude_desktop_config.json. Keep existing servers. Add any arguments, credentials, and permissions required by the maintainer; this template has not been install-tested.
{
"mcpServers": {
"dev-pagelens-pagewatch": {
"command": "npx",
"args": [
"-y",
"pagewatch-mcp"
]
}
}
}Restart Claude Desktop completely for changes to take effect. Confirm the server appears connected in the client’s tool list, then try a read-only example from its documentation.
Claude Desktop setup referencepagewatch-mcpnpmPagewatch works with any MCP-compatible client. Copy the config snippet from the Configuration section above and add it to the file shown for your client, then restart the application.
~/Library/Application Support/Claude/claude_desktop_config.jsonRestart Claude Desktop completely for changes to take effect.~/.cursor/mcp.jsonRestart Cursor for changes to take effect..vscode/mcp.jsonReload VS Code window for changes to take effect.~/.codeium/windsurf/mcp_config.jsonRestart Windsurf for changes to take effect..mcp.jsonSave at the project root, then start Claude Code in that project and review the MCP server approval prompt. Keep real credentials out of shared files.