Stagehand / Browserbase
browser_engine: stagehand — drives a cloud browser hosted on Browserbase through the Stagehand MCP server. Instead of low-level primitives, it exposes AI-driven, natural-language tools: navigate, act, observe, extract, plus session lifecycle (start/end).
Reference: Stagehand MCP setup.
When to use it
Section titled “When to use it”- You want browser automation without installing or maintaining local browsers.
- You want resilient, AI-driven steps (
act("click the Sign in button"),extract("the order total")) that tolerate small DOM changes. - You already have a Browserbase account.
Prerequisites
Section titled “Prerequisites”- A Browserbase account.
- A Browserbase API key and Project ID from the dashboard.
Configuration
Section titled “Configuration”This plugin uses the hosted (HTTP) Browserbase MCP. /qa-catalog:init writes this mcpServers block into each project-level browser agent:
mcpServers: - browserbase: type: http url: "https://mcp.browserbase.com/mcp?browserbaseApiKey=${BROWSERBASE_API_KEY}"Export your key before launching Claude Code so the ${BROWSERBASE_API_KEY} placeholder resolves:
export BROWSERBASE_API_KEY="bb_live_..."export BROWSERBASE_PROJECT_ID="..." # used by some optional query paramsYou can also register it from the CLI instead of the inline block:
claude mcp add --transport http browserbase \ "https://mcp.browserbase.com/mcp?browserbaseApiKey=$BROWSERBASE_API_KEY"Optional query parameters
Section titled “Optional query parameters”| Param | Purpose |
|---|---|
| modelName | LLM driving act/extract (default google/gemini-2.5-flash-lite). |
| modelApiKey | Your own model key if you bring your own LLM. |
| proxies=true | Route through Browserbase proxies. |
| advancedStealth=true | Anti-bot stealth mode. |
| keepAlive=true | Keep sessions warm between calls. |
Isolation & parallelism
Section titled “Isolation & parallelism”Each spawn opens its own Browserbase cloud session, so isolation is handled server-side — no shared local profile to worry about. Mind your Browserbase plan’s concurrent-session limit when setting parallel_agents / parallel_test_runners.
Tool-name differences
Section titled “Tool-name differences”On this engine the agent uses navigate, act, observe, and extract instead of the browser_* primitives. Map app actions to natural-language act calls and read state with observe / extract. See the capability map.
Settings that do not apply
Section titled “Settings that do not apply”browser_channelis ignored — the cloud chooses the browser.browser_headlessis ignored — cloud sessions are headless by nature.