cdp-browser

n8n community node for a self-hosted Remote Chromium CDP Automation REST API - full tab, JS, input, form, capture, cookie, storage, network, download, geolocation, emulation, dialog and permission control.

Package Information

Released: this month
Downloads: 0 weekly / 0 monthly
Latest Version: 1.0.0
Author: Your Name

Available Nodes

Documentation generating…

CDP Browser Automation icon
CDP Browser Automation
Control a remote Chromium instance via your CDP Automation API server

Documentation

n8n-nodes-cdp-browser

An n8n community node for a self-hosted Remote Chromium CDP Automation
REST API (a FastAPI wrapper around --remote-debugging-port). It exposes
the whole API surface as one node with 16 resources / ~45 operations:

  • Diagnostics — health, version
  • Tab — list / get / create / close / close all / activate / attach /
    detach / is-attached
  • Navigation — navigate
  • HTML / JavaScript — get HTML, evaluate JS
  • Mouse — click (selector or point), click at point, double click,
    move, drag, scroll
  • Keyboard — type, key press
  • Form — set value, select, check/uncheck, submit
  • Page Capture — screenshot (PNG/JPEG, full page, by selector), PDF
  • Cookie — get, set, delete
  • Storage — get/set/delete local or session storage, clear origin
  • Network — log, block, headers, throttle, cache, auth, mock
  • Download — configure, list, clear
  • Geolocation — set/clear
  • Emulation — set/clear device viewport & UA
  • Dialog — get pending, set mode, respond
  • Permission — grant, reset

Install

In n8n: Settings → Community Nodes → Install, package name
n8n-nodes-cdp-browser.

Or manually, from source:

npm install
npm run build
npm link
# in your n8n install / ~/.n8n/nodes:
npm link n8n-nodes-cdp-browser

Credentials

Create a CDP Browser Automation API credential:

  • Base URL — e.g. http://localhost:8000
  • API Key — only needed if your server was started with API_KEY set
    (sent as X-API-Key)

Notes

  • Target ID. Most operations need the tab's target_id. Get it from
    Tab → List or Tab → Create. A few read-oriented operations (Get
    HTML, Evaluate, Screenshot, PDF) fall back to the first open tab when you
    leave Target ID empty.
  • Selectors vs. coordinates. Mouse → Click lets you pick either; never
    set both.
  • Screenshot / PDF are returned as n8n binary data (default output
    field: data) so they chain straight into a Write Binary File, Upload,
    or HTTP Request node.
  • Close All Tabs requires the Confirm toggle — a deliberate guard
    against wiping every tab by accident.

License

MIT

Discussion