by Browserbase
Stagehand is Browserbase's SDK for browser agents, with natural language selectors, self-healing actions and caching. It runs on Browserbase's cloud browsers.
APIUsage-based
Step 1: Sign up for a Browserbase account and copy your API key from the dashboard.
Step 2: Try a template in the Playground and watch it in Live View.
Step 3: Follow the Stagehand quickstart to run browser actions from your code.
Step 4: Inspect finished runs in the Session Inspector.
Step 5: Deploy the agent as a Browserbase Function if you want it scheduled or callable by API.
No. You can control sessions with Playwright, Puppeteer or Selenium, but the docs recommend Stagehand for AI-native workflows.
Use Live View during a run, then open the Session Inspector to view recordings and logs.
The docs describe Agent Identity for giving agents a verifiable identity and handling authentication on the web.
Deploy it with Browserbase Functions, which run on a schedule or on demand and can be invoked by API.
BrowserbaseBrowserbase is cloud infrastructure for AI agents that use the web, with headless browsers, Search and Fetch APIs, and agent identity tools. Agents can log in, navigate and act on sites like a person.
Braintrust EvalsBraintrust Evals is an AI-native product for evaluation SDK and platform for testing prompts, models, scorers, datasets, and AI application regressions.
FirecrawlFirecrawl is a web data API that searches, scrapes and interacts with web pages and returns clean, LLM-ready data. It is open source and works with AI agents through MCP and a CLI.
Pl@ntNet APIThe Pl@ntNet API identifies plant species from photos through a REST API. It covers 87,000+ species and has a free plan of 500 identifications per day.