
Parallel Browser MCP
io.github.etairlPublished by GitHub account etairlv0.1.14Updated Oct 24, 2026
MCP server for parallel browser automation across multiple providers.
The publisher's own server: Parallel Search MCP
Overview
Lets an assistant run and control several browser sessions in parallel across local Playwright or cloud browser providers.
- What you can do
- Exposes a numeric session model over MCP so one client can create and control multiple browser sessions at once. Session tools include start_session, close_session, close_all_sessions and get_sessions; every browser_* tool takes a sessionId. Browser tools cover navigation, clicking, form filling, screenshots, snapshots, page structure, locators, keyboard and mouse actions, file upload, waiting and JavaScript evaluation. Providers are abstracted: local Playwright Chromium, Browserbase, Anchor Browser and Cloudflare Browser Run.
- When to use it
- Useful when an assistant needs to drive real web pages, run several browsing tasks concurrently, or test the same flow against different browser providers. Also suited to agent workflows that need screenshots, page snapshots or form interaction rather than plain text fetching.
- Requirements
- Runs locally over stdio, typically via npx parallel-browser-mcp@latest, so Node.js is needed. The local playwright provider needs no credentials; browserbase needs BROWSERBASE_API_KEY plus a project ID, anchor needs ANCHOR_API_KEY, and cloudflare needs CLOUDFLARE_API_TOKEN and CLOUDFLARE_ACCOUNT_ID. Provider settings come from BROWSER_MCP_CONFIG or BROWSER_MCP_CONFIG_PATH. Optional stealth mode requires the cloakbrowser package and downloads a large Chromium binary.
Installation
In SourceWeft
- Open Parallel Browser MCP in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.
Other MCP clients
Follow the launch instructions in the repository.
README
parallel-browser-mcp
parallel-browser-mcp is an MCP server for parallel browser automation. It exposes a numeric session model over MCP so one client can create and control multiple browser sessions at the same time across multiple browser providers.
Supported providers:
playwrightfor local Chromiumbrowserbasevia@browserbasehq/sdkanchorviaanchorbrowsercloudflarevia Cloudflare Browser Run
Each browser session gets a numeric ID like 1, 2, 3, and every browser_* tool accepts a sessionId.
Features
- Multiple concurrent browser sessions in memory
- Provider abstraction shared across Browserbase, Anchor Browser, Cloudflare Browser Run, and local Playwright
- MCP session tools:
start_sessionclose_sessionclose_all_sessionsget_sessions
- Browser tools:
browser_navigatebrowser_go_backbrowser_clickbrowser_fillbrowser_fill_formbrowser_screenshotbrowser_snapshotbrowser_hoverbrowser_dragbrowser_select_optionbrowser_generate_locatorbrowser_get_page_structurebrowser_evaluatebrowser_keyboard_pressbrowser_keyboard_typebrowser_mouse_movebrowser_mouse_click_xybrowser_mouse_dragbrowser_upload_filebrowser_wait_for_selectorbrowser_wait_for_timeout
Quick Start
Run locally over stdio:
Run it as an npm package CLI:
Configuration
Provider-specific settings are configured at the MCP server configuration level, not per tool call.
The server reads config in this order:
BROWSER_MCP_CONFIGBROWSER_MCP_CONFIG_PATH- direct env defaults
- built-in defaults
Recommended config shape:
Stealth Chromium via CloakBrowser
The playwright provider can optionally launch CloakBrowser instead of vanilla Chromium for sessions that need to bypass bot detection. Enable it per-config or via env:
cloakbrowser is an optional peer — install it only when you need stealth:
The CloakBrowser binary (~200MB stealth Chromium) is downloaded automatically on the first session launch. Existing launchOptions / contextOptions continue to apply, and the rest of the provider behaves identically to standard Playwright.
Required credentials by provider:
playwright: nonebrowserbase:BROWSERBASE_API_KEY, plus aprojectIdin config orBROWSERBASE_PROJECT_IDanchor:ANCHOR_API_KEYcloudflare:CLOUDFLARE_API_TOKEN,CLOUDFLARE_ACCOUNT_ID
Optional env defaults:
BROWSERBASE_PROJECT_IDBROWSERBASE_KEEP_ALIVEBROWSERBASE_CONTEXT_IDBROWSERBASE_PERSISTPLAYWRIGHT_STORAGE_STATE_PATHPLAYWRIGHT_EXECUTABLE_PATHPLAYWRIGHT_CHANNELPLAYWRIGHT_USE_CLOAKBROWSER(trueto launch stealth Chromium via CloakBrowser; requiresnpm install cloakbrowser)
Installation
Use the standard config below in any MCP client that supports stdio:
Claude Code
Use the Claude Code CLI to add the server:
If you need provider configuration, add the environment variables in your Claude MCP config using the standard config above.
Claude Desktop
Follow the Claude Desktop MCP install flow and use the standard config above in the local MCP configuration file.
Codex
Use the Codex CLI:
Or add this to ~/.codex/config.toml:
Copilot
Use the Copilot CLI interactive flow:
Or add this to ~/.copilot/mcp-config.json:
Cursor
Go to Cursor Settings -> MCP -> Add new MCP Server, then use:
- command:
npx - args:
parallel-browser-mcp@latest
Or paste the standard config above into the MCP config editor.
Gemini
Add the server to .gemini/settings.json:
VS Code
Use the MCP install flow in VS Code with the standard config above, or install with the VS Code CLI:
Example Flow
- Call
start_sessionwith{ "provider": "playwright" } - Read the returned session
id - Call
browser_navigatewith{ "sessionId": 1, "url": "https://example.com" } - Call any additional
browser_*tool with the samesessionId - Call
close_sessionwhen done
Development
Publishing
This repo is set up to publish as an npm package:
- the CLI entrypoint is
parallel-browser-mcp - production builds exclude tests and smoke scripts
- the published package only includes
dist,README.md, and.env.example
Before publishing:
GitHub Actions publishing:
.github/workflows/publish.ymlpublishes to npm on GitHub release publication or manual dispatch- set the
NPM_TOKENrepository secret before using the publish workflow
Examples
examples/localcontains a standalone npm package that connects toparallel-browser-mcpwith@langchain/mcp-adaptersand runs a LangChain agent against the local Playwright provider.examples/browserbasecontains a standalone npm package that connects LangChain to the published MCP server with Browserbase config and prompts the agent to usebrowser_screenshot.examples/anchorcontains a standalone npm package that connects LangChain to the published MCP server with Anchor config and prompts the agent to usebrowser_snapshot.examples/cloudflarecontains a standalone npm package that connects LangChain to the published MCP server with Cloudflare Browser Run config and prompts the agent to usebrowser_snapshot.- The root
.npmignoreexcludes the fullexamplesdirectory from npm publishing.
Testing
The repo includes:
- unit coverage for config loading, providers, registry behavior, session tools, and representative browser tools
- a local Playwright smoke script in
src/smoke/localSmoke.ts
Notes
start_sessionis intentionally small. Provider-specific behavior belongs in MCP configuration, not tool inputs.- The server logs to stderr so stdout stays clean for MCP JSON-RPC traffic.
- Browserbase and Anchor Browser are normalized to Playwright page operations after connection, so the browser tools stay provider-agnostic.
Source: README.md at commit bde38ec
Tools
0Version history
1- v0.1.14LatestSep 16, 2026
Score
50% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 days25 pts
Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.
- Auth declared10 pts
Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
Other Parallel integrations

Parallel Search MCP
ai.parallelOfficial publisher
A hosted web search and page-fetch server that lets an assistant find current information and read web pages, with free anonymous access.

Parallel Task MCP
ai.parallelOfficial publisher
A hosted Parallel service that lets an assistant start deep research and data-enrichment jobs, check their progress, and retrieve sourced results as Markdown.

CLIO Parallel Sort
Published by GitHub account iowarp
CLIO Kit's parallel-sort server lets an assistant sort, filter and summarize large log files by timestamp, level or keyword.

Parallel
com.mcparmory
Lets an assistant search the web, extract page content, and run distributed automation tasks through the Parallel API.