Skip to content

Cloudflare Browser Run Review

Cloudflare Browser Run is an AI tool designed to programmatically control and interact with headless browser instances on Cloudflare's global network.

shipped Jun 8, 2026aifreemium
Cloudflare Browser Run - AI tool
1Rebuilt on Cloudflare Containers on May 13, 2026, for enhanced performance.
2Increased concurrency limits by 4x, supporting up to 120 concurrent browsers for Workers Paid plans.
3Achieved over 50% faster response times for Quick Actions after the May 2026 rebuild.
4Renamed from 'Browser Rendering' to 'Browser Run' on April 15, 2026, to reflect expanded capabilities.

Stork Quadrant

Becomes the API· 56/100

Replaceable as a UI, but kept alive as the API the agents call.

Cloudflare Browser Run is a headless browser execution layer running on Cloudflare's global edge network. The LLM can write the script; it cannot run it at scale across 300 PoPs with built-in bot bypass, IP reputation, and zero cold-start latency. The physical infrastructure and coordination with Cloudflare's existing network stack are real moats — this isn't a wrapper, it's a runtime.

Claude Sonnet 4.6, scored 2026-06-08

Defensibility · 33/100

  • Physical-world coupling
  • Regulatory moat
  • Network liquidity
  • Proprietary refreshing data
  • High-trust catastrophic workflows
  • Multi-party coordination
  • Brand / community / taste

An LLM alone could replace

  • Summarize or extract content from a webpage given a URL
  • Generate browser automation scripts (Puppeteer, Playwright)
  • Parse and analyze HTML structure or DOM content
  • Write code to scrape or interact with web elements

Agent-Readiness · 85/100

  • Verified MCPStork MCP listing: cloudflare-mcp (confirmed)
  • Listed on agent surfacesanthropic_directory, cursor + Stork:cloudflare-mcp
  • Usage-based pricingpricing page heuristic match: https://cloudflare.com/pricing
  • Headless agent auth
  • Public OpenAPIhttps://docs.cloudflare.com/openapi.json
  • Active changeloghttps://cloudflare.com/blog (2026-06-05)
  • llms.txthttps://cloudflare.com/llms.txt

How to defend

Double down on the agent-callable API angle — become the browser execution primitive that every AI agent framework calls when it needs to render, scrape, or interact with the live web. Own the 'browser as a tool' slot in the MCP/function-calling ecosystem before someone else does.

  • Expose API-key auth with a self-serve sandbox tier; remove sales-call gates (+15).

Cloudflare Browser Run at a Glance

Best For
Developers, businesses, and organizations looking for secure and scalable internet solutions.
Pricing
Subscription SaaS
Key Features
Rebuilt on Cloudflare Containers on May 13, 2026, for enhanced performance. · Increased concurrency limits by 4x, supporting up to 120 concurrent browsers for Workers Paid plans. · Achieved over 50% faster response times for Quick Actions after the May 2026 rebuild.
Alternatives
Browserless, Apify, ScrapingBee, Bright Data (Web Scraper IDE)

About Cloudflare Browser Run

Business Model
Subscription SaaS
Headquarters
San Francisco, USA
Founded
2009
Team Size
1001-5000
Funding
Public
Total Raised
$1.75 billion
Target Audience
Developers, businesses, and organizations looking for secure and scalable internet solutions.

Connect

</>Embed "Featured on Stork" Badge
Badge previewBadge preview light
<a href="https://www.stork.ai/en/cloudflare-browser-run" target="_blank" rel="noopener noreferrer"><img src="https://www.stork.ai/api/badge/cloudflare-browser-run?style=dark" alt="Cloudflare Browser Run - Featured on Stork.ai" height="36" /></a>
[![Cloudflare Browser Run - Featured on Stork.ai](https://www.stork.ai/api/badge/cloudflare-browser-run?style=dark)](https://www.stork.ai/en/cloudflare-browser-run)

overview

What is Cloudflare Browser Run?

Cloudflare Browser Run is a browser automation and AI agent browsing tool developed by Cloudflare that enables developers and AI agents to programmatically control and interact with headless browser instances. It provides access to headless Chrome instances controllable via standard automation libraries like Puppeteer, Playwright, or the Chrome DevTools Protocol (CDP). This tool allows for the automation of browser tasks without a graphical user interface, leveraging Cloudflare's global network for execution. Its primary applications include end-to-end testing of web applications, web scraping and data extraction from dynamic webpages, secure investigation of suspicious URLs in isolated environments, and content generation such as screenshots and PDFs. A significant recent emphasis has been on enabling AI agents to interact with the web at scale, serving as the 'eyes and hands' for AI models to navigate, fill forms, and extract data where traditional APIs are absent. Browser Run offers two main integration methods: Quick Actions for stateless tasks via REST API or Workers binding, and Browser Sessions for direct, full browser control using Puppeteer, Playwright, CDP, or Cloudflare's Stagehand, deployable within Cloudflare Workers or connectable from any environment.

quick facts

Quick Facts

AttributeValue
DeveloperCloudflare
Business ModelFreemium
PricingFreemium
PlatformsAPI, Web (via Cloudflare Workers)
API AvailableYes
IntegrationsPuppeteer, Playwright, Chrome DevTools Protocol (CDP), Stagehand, Cloudflare Workers, WebMCP
Founded2009
HQSan Francisco, USA
FundingPublic, $1.75 billion

features

Key Features of Cloudflare Browser Run

Cloudflare Browser Run provides a comprehensive suite of features for headless browser automation and AI agent interaction, leveraging Cloudflare's global infrastructure. These capabilities are designed to offer granular control, high performance, and robust debugging tools for complex web workflows.

  • 1Headless Chrome Instances: Access to fully functional headless Chrome browsers for automation.
  • 2Multiple Control Libraries: Support for Puppeteer, Playwright, Chrome DevTools Protocol (CDP), and Stagehand for flexible scripting.
  • 3Quick Actions: Simplified API endpoints for common tasks like screenshots, PDF generation, and basic content extraction.
  • 4Browser Sessions: Full, persistent browser control for multi-step workflows and complex interactions.
  • 5Live View: Real-time visual monitoring and interaction with remote browser sessions for debugging.
  • 6Human in the Loop: Capability for human intervention during automated sessions to handle CAPTCHAs or unexpected scenarios.
  • 7Session Recordings: Comprehensive capture of browser sessions, including DOM changes and user interactions, for post-mortem analysis.
  • 8WebMCP (Model Context Protocol) Support: Enables AI agents to discover and execute structured tools on websites for more reliable navigation.
  • 9Cloudflare Workers Integration: Seamless deployment and execution of browser automation scripts directly on Cloudflare's edge network.
  • 10WebGL Support: Enhanced browser rendering capabilities for applications requiring advanced graphics.

use cases

Who Should Use Cloudflare Browser Run?

Cloudflare Browser Run is designed for a diverse range of users requiring programmatic interaction with web content, from individual developers to large enterprises and specialized teams. Its capabilities are particularly beneficial for automating web workflows, extracting data, and enabling advanced AI agent functionality.

  • 1Individual Developers: For automating web tasks, performing end-to-end testing of web applications, and conducting web scraping for personal projects.
  • 2Businesses (Small to Large Enterprises): To securely investigate suspicious URLs in isolated environments, generate content (e.g., marketing materials from web pages), and build full-stack applications requiring browser interaction.
  • 3AI Agent Developers: To provide AI agents with the ability to navigate websites, fill forms, extract specific data, and perform multi-step workflows, especially where traditional APIs are unavailable.
  • 4Security Teams: For analyzing web content and identifying potential threats within a controlled, isolated browser environment.
  • 5Web Application Developers: For automating user flows and ensuring the functionality and performance of web applications through comprehensive end-to-end testing.

pricing

Cloudflare Browser Run Pricing & Plans

Cloudflare Browser Run operates on a freemium model, offering a free tier with specific usage limits and paid plans for increased capacity and advanced features. The free tier allows users to get started with browser automation and quick actions. For higher usage, such as increased concurrency limits (e.g., 120 concurrent browsers, up from 30, for Workers Paid plans as of May 13, 2026) and faster instance rates (1 per second), users can subscribe to Cloudflare Workers Paid plans. Specific pricing for these paid plans is integrated within the broader Cloudflare Workers platform pricing structure, which typically involves a base subscription plus usage-based fees for compute and other services. The service was rebuilt on Cloudflare Containers in May 2026, leading to higher usage limits and improved performance across all tiers.

  • 1Free Tier: Includes basic browser automation and quick actions with defined usage limits.
  • 2Workers Paid Plans: Offers increased concurrency (e.g., 120 concurrent browsers), faster response times, and higher instance rates, integrated with Cloudflare Workers platform pricing.

competitors

Cloudflare Browser Run vs Competitors

Cloudflare Browser Run competes in the headless browser automation and web scraping market against several established platforms. Its key differentiators often stem from its deep integration with Cloudflare's global edge network and its specific focus on enabling AI agents.

1
Browserless

Browserless provides a highly scalable and reliable headless browser as a service, focusing on raw Puppeteer and Playwright execution without managing infrastructure.

Similar to Cloudflare Browser Run, Browserless offers headless Chrome/Playwright for automation and scraping. However, Cloudflare Browser Run integrates more deeply with its edge network and AI ecosystem (Workers AI, AI agents), offering AI-powered data extraction and agent browsing capabilities directly, whereas Browserless primarily provides the underlying browser infrastructure.

2

Apify is a comprehensive platform for web scraping, data extraction, and browser automation, offering a wide range of pre-built actors and custom development tools.

Apify competes with Cloudflare Browser Run in web scraping and automation, providing a more extensive marketplace of 'Actors' for various tasks. While Apify can be used to build AI-driven agents, Cloudflare Browser Run is more tightly integrated into Cloudflare's broader AI infrastructure (Workers AI, Agents SDK) and emphasizes edge deployment and AI agent control via protocols like CDP and WebMCP.

3
ScrapingBee

ScrapingBee is a web scraping API that handles headless browsers and proxies, simplifying the process of extracting data from dynamic websites.

ScrapingBee directly competes with Cloudflare Browser Run's web scraping and data extraction features, particularly its 'Quick Actions' for screenshots and content extraction. Cloudflare Browser Run, however, offers more granular control over headless browser sessions via Puppeteer/Playwright/CDP and is explicitly positioned as a platform for AI agents to interact with the web at scale, including AI-powered data extraction with natural language prompts.

4
Bright Data (Web Scraper IDE)

Bright Data's Web Scraper IDE provides a robust, cloud-based environment for developing, deploying, and managing complex web scraping projects with integrated proxy networks and data parsing tools.

Bright Data's Web Scraper IDE offers a powerful solution for web scraping and data extraction, similar to Cloudflare Browser Run's capabilities. While Bright Data provides extensive proxy networks and a visual IDE for scraper development, Cloudflare Browser Run leverages its global edge network for low-latency browser execution and is specifically designed to be the 'browser for AI agents,' offering direct integration with AI models and agent control protocols.

Frequently Asked Questions

+What is Cloudflare Browser Run?

Cloudflare Browser Run is a browser automation and AI agent browsing tool developed by Cloudflare that enables developers and AI agents to programmatically control and interact with headless browser instances. It provides access to headless Chrome instances controllable via standard automation libraries like Puppeteer, Playwright, or the Chrome DevTools Protocol (CDP).

+Is Cloudflare Browser Run free?

Yes, Cloudflare Browser Run offers a freemium model, including a free tier with specific usage limits for basic browser automation and quick actions. For increased concurrency and advanced features, users can subscribe to Cloudflare Workers Paid plans, which integrate Browser Run's capabilities.

+What are the main features of Cloudflare Browser Run?

Key features include headless Chrome instances, support for Puppeteer, Playwright, CDP, and Stagehand, Quick Actions for simple tasks, full Browser Sessions for complex workflows, Live View for real-time debugging, Human in the Loop for intervention, Session Recordings, and WebMCP support for AI agents. It also integrates with Cloudflare Workers and supports WebGL.

+Who should use Cloudflare Browser Run?

Cloudflare Browser Run is suitable for individual developers, small to large businesses, AI agent developers, security teams, and web application developers. It is used for end-to-end testing, web scraping, secure URL investigation, content generation, and enabling AI agents to interact with the web at scale.

+How does Cloudflare Browser Run compare to alternatives?

Cloudflare Browser Run differentiates itself from competitors like Browserless, Apify, ScrapingBee, and Bright Data by its deep integration with Cloudflare's global edge network, its specific focus on enabling AI agents through protocols like CDP and WebMCP, and its direct integration with the Cloudflare Workers ecosystem for serverless deployment and execution.

For builders

This page is doing a job for someone else’s tool.

AI agents read it. Buyers find it. Backlinks accrue. Your tool can have one too — live in 24 hours, indexed by Claude, ChatGPT, and Perplexity, queryable via MCP.