Parallel Browser MCP
Run browser automation through one MCP surface while switching or scaling across Browserbase, Anchor, and other configured providers in parallel.
Overview
io.github.etairl/parallel-browser-mcp is an MCP server for the Build phase that runs parallel browser automation across multiple third-party browser providers from one config.
What is this MCP server?
- Parallel Browser MCP npm package parallel-browser-mcp v0.1.8
- Multi-provider browser automation behind one MCP server
- Config via BROWSER_MCP_CONFIG inline JSON or BROWSER_MCP_CONFIG_PATH file
- Supports Browserbase (API key + project ID) and Anchor (ANCHOR_API_KEY) providers
- stdio transport with npx runtime hint
- Version 0.1.8
- npm identifier parallel-browser-mcp
- stdio transport with npx runtimeHint
What problem does it solve?
Agent browser tasks stall when each provider needs a different SDK and you cannot run sessions in parallel across vendors.
Who is it for?
Builders running agent-led web flows who want Browserbase or Anchor behind a single parallel MCP layer.
Skip if: Local-only headless Chrome with zero cloud API keys or teams that only need a single-provider Playwright script.
What do I get? / Deliverables
One MCP registration lets your agent drive browsers on configured providers with shared JSON or env-based configuration.
- Unified MCP tools for multi-provider browser control
- Parallel automation capacity across configured vendors
- Centralized browser MCP configuration file or env block
Recommended MCP Servers
Journey fit
Browser MCP integrations are wired during Build when agents need to navigate real sites, scrape flows, or exercise your app in hosted browsers. Integrations is the shelf for MCP servers that connect agents to external runtimes and third-party browser infrastructure.
How it compares
Multi-provider browser MCP integration, not a curated skill marketplace or static web scraper CLI.
Common Questions / FAQ
Who is parallel-browser-mcp for?
Developers using Claude Code, Cursor, or similar agents who need cloud browser sessions from more than one vendor through MCP.
When should I use parallel-browser-mcp?
Use it when building or testing web flows that require hosted browsers, failover between providers, or parallel agent-driven navigation.
How do I add parallel-browser-mcp to my agent?
Register stdio MCP with npx parallel-browser-mcp, set BROWSER_MCP_CONFIG or BROWSER_MCP_CONFIG_PATH, and add provider keys such as BROWSERBASE_API_KEY or ANCHOR_API_KEY.