REMOTE BROWSER / LEARNING CENTER
Understand the browser.
Build better agents.
Learn how remote browsers work. Explore CDP, sessions, persistent profiles, agent integrations, deployment comparisons, and reproducible benchmark methods.

THE FOUNDATION
What is a remote browser?
A remote browser runs separately from the program controlling it. Your agent or automation client sends commands over a network connection; the browser loads websites, runs JavaScript, and holds the active browsing state. Hosted infrastructure adds session management, access controls, profiles, and debugging tools around that runtime.
Start with the concepts, choose your integration, then measure your own workflow. The collections below bring together focused guides and product documentation.
01 / 6 RESOURCES
Guides
Understand remote browsers, CDP, sessions, and persistent profiles.
What is remote CDP?
Learn where CDP fits between your automation client and a network-accessible Chromium browser.
Read guide GuideWhat is a browser session?
Separate the lifetime of a connection, a browser process, and an authenticated website session.
Read guide GuideWhat is a persistent browser profile?
Understand saved browser state, profile ownership, expiry, and safe reuse between tasks.
Read guide GuideWhat is a remote browser?
An introduction to hosted Chromium, CDP connections, and browser runtime choices.
Read guide GuideConnect Playwright to remote Chrome
Attach to an existing Chromium browser and troubleshoot the connection.
Read guide GuideConnect Puppeteer to a remote browser
Use a browser WebSocket endpoint with Puppeteer.
Read guide02 / 6 RESOURCES
Infrastructure
Design browser lifecycles, isolation, authentication, and network routing.
Browser session isolation
Distinguish browser contexts from process, storage, and network isolation.
Read guide GuideHow remote browser sessions work
Design allocation, readiness, attachment, reconnects, and cleanup as an explicit lifecycle.
Read guide DocumentationRemote Browser architecture
Explore the control plane, session pods, CDP gateway, viewer, and storage.
Read documentation DocumentationBrowser authentication and profile management
Configure persistent profiles and reuse authenticated state.
Read documentation DocumentationBrowser proxy architecture
Choose proxy routing for your browser session and validate egress.
Read documentation DocumentationRemote CDP security
Review session access, credentials, recordings, and storage boundaries.
Read documentation03 / 8 RESOURCES
Browser agents
Connect agent frameworks, coding agents, Playwright, and Puppeteer to remote Chromium.
Playwright MCP with a remote browser
Connect the MCP server to an existing Chromium endpoint and verify a read-only task.
Read integration IntegrationBrowser Use with remote CDP
Understand how the agent layer connects to hosted Chromium.
Read integration Integrationagent-browser with remote CDP
Explore the CDP connection and browser runtime responsibilities.
Read integration IntegrationGive Codex and Claude a remote browser
Install the Remote Browser Agent skill and configure its CDP endpoint.
Read integration IntegrationHermes with Remote Browser
Install and configure the Hermes integration.
Read integration IntegrationOpenClaw with Remote Browser
Connect an OpenClaw browser profile to remote Chromium.
Read integration IntegrationPlaywright over remote CDP
Use the existing product guide for a Playwright-based automation client.
Read integration IntegrationPlaywright, Puppeteer, and Selenium setup
Find framework-specific connection instructions in the documentation.
Read integration04 / 6 RESOURCES
Comparisons
Compare deployment models, browser modes, agent tools, and runtime responsibilities.
Local browser vs remote browser automation
Compare deployment, latency, state, debugging, and operational ownership.
Read comparison ComparisonHeadless browser vs remote browser
Understand why display mode and deployment location are independent choices.
Read comparison ComparisonBrowser agent vs browser infrastructure
Separate the decision loop, harness, connection protocol, and browser runtime.
Read comparison ComparisonBrowser Use vs agent-browser
Compare Python agent workflows and command-oriented browser automation.
Read comparison ComparisonBrowser as a service vs self-hosted Playwright
Evaluate infrastructure ownership and the cost of operating browser workers.
Read comparison ComparisonRemote Browser vs Browserbase
Evaluate product fit and browser infrastructure responsibilities.
Read comparison05 / 4 RESOURCES
Benchmarks
Measure CDP attachment, navigation, task success, and browser resource use.
Benchmark local and remote CDP latency
Run a timing script that records attachment, navigation, failures, and environment metadata.
Read methodology MethodologyRemote Browser benchmark methodology
Define phases, sample size, percentiles, and disclosure requirements.
Read methodology GuideMulti-session browser resource usage
Plan measurements for CPU, memory, concurrency, and browser session pressure.
Read guide GuideMeasure billable browser-hours
Understand how browser lifetime affects the cost of a completed workflow.
Read guide