Integration

Browserbase integration

Browserbase headless browser API. Start cloud browser sessions with proxies, stealth settings and a chosen region, list and inspect them, request their release, read their console and network logs, pull the rrweb recording, download the files a session produced, upload a file into a running session, and manage the contexts and extensions that persist state between sessions. It also exposes a one shot page fetch and a web search that need no session, plus project level usage. Driving the browser itself happens over the returned WebSocket URL with a browser automation client, not over this API.

Connect Browserbase How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Browserbase.Built from the 18 Browserbase tools below, on the connection you add once.

Common Browserbase use cases

finding sessions left running
finding candidate pages before fetching them
finding the project id to create sessions in
reading an article as markdown for a summary
removing an extension that is no longer used
watching consumption before hitting a plan limit
replaying what the browser did during a failed run
clearing a stored login after an account is rotated
confirming a context still exists before reusing it
showing a user the browser while an automation runs

What you can do with Browserbase

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Session, 4 operations

create_sessionCreate sessionPOSTStarts a cloud browser session and returns the URL an automation client connects to.
get_sessionGet sessionGETReturns one session with its status, proxy usage and connection URLs.
list_sessionsList sessionsGETLists browser sessions, optionally filtered by status or by user metadata.
update_sessionUpdate sessionPOSTEnds a running session by requesting its release.

Context, 3 operations

create_contextCreate contextPOSTCreates a reusable browser context that carries cookies and local storage between sessions.
delete_contextDelete contextDELETEDeletes a context and the browser state it stored.
get_contextGet contextGETReturns a context with its name and project.

Extension, 2 operations

delete_extensionDelete extensionDELETEDeletes an uploaded extension.
upload_extensionUpload extensionPOSTUploads a zipped Chrome extension that sessions can load.

Page, 1 operation

fetch_pageFetch pagePOSTFetches a URL through Browserbase and returns its content, optionally as markdown or structured JSON.

Project, 1 operation

list_projectsList projectsGETLists projects with their default timeout and concurrency limit.

Project usage, 1 operation

get_project_usageGet project usageGETReturns the browser minutes and proxy bytes consumed by a project.

Search result, 1 operation

search_webSearch webPOSTRuns a web search and returns ranked results with titles and URLs.

Session debug, 1 operation

get_session_live_urlsGet session live urlsGETReturns the live debugger URLs that let a human watch or take over a running session.

Session download, 1 operation

download_session_filesDownload session filesGETDownloads everything the browser saved during a session as a single zip archive.

Session log, 1 operation

get_session_logsGet session logsGETReturns the browser protocol log of a session, request and response by request.

Session recording, 1 operation

get_session_recordingGet session recordingGETReturns the recorded rrweb events of a session for replay.

Session upload, 1 operation

upload_session_fileUpload session filePOSTPuts a file into a running session so the page can attach it to a file input.

Use Browserbase from Claude Code, Cursor or Codex

Connect Browserbase in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Browserbase operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Browserbase

Create your workspace
Sign up free. Nothing to install.
Connect Browserbase once
Add your Browserbase connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.

Browserbase integration FAQ

What can you do with the Browserbase integration?

LiveContext exposes 18 operations on Browserbase, covering session, context, extension and page. Common uses include finding sessions left running, finding candidate pages before fetching them and finding the project id to create sessions in.