# Stagehand — card for AI agents

Open-source SDK from Browserbase for building browser agents that mix natural-language actions (act, extract, observe, agent) with code, available in TypeScript, Python and Go and runnable locally or on Browserbase cloud browsers.

Maker: Browserbase · category: Browser and computer-use agents · open source (MIT)

## Price
- Browserbase offers browser automation and agent infrastructure. Free tier: 1 browser hour ($5 in model tokens), 3 concurrent browsers, limited API calls. Developer $20/month: 100 browser hours then $0.12/hr overage, 25 concurrent browsers, usage-based overages on Search ($5/1k) and Fetch ($1-7/1k) calls. Startup $99/month: 500 browser hours then $0.10/hr, 100 concurrent browsers, higher limits with per-unit overages. Enterprise: custom pricing with 250+ browsers and usage-based billing. Model tokens billed separately on all plans.
- Free tier: yes
- Paid from $20/month with monthly billing
- Pricing page: https://www.browserbase.com/pricing
- Checked: 2026-10-15

## Install
- npm: `npm install @browserbasehq/stagehand`
- PyPI: `pip install stagehand`

## Facts
- Interfaces: API
- Bring your own API key: yes
- Repository: https://github.com/browserbase/stagehand

## Numbers (AgentGid data)
- Gid Score: 57 (rank #34) (public popularity and activity, not quality)
- Downloads last 7 days: 2,874,961
- GitHub stars: 26,096

## Details (plan limits, live status, safety; from AgentGid's data)
Stagehand by Browserbase. Category: Browser and computer-use agents. Type: browser automation library. Open-source SDK from Browserbase for building browser agents that mix natural-language actions (act, extract, observe, agent) with code, available in TypeScript, Python and Go and runnable locally or on Browserbase cloud browsers. Open source (MIT) Pricing (checked 2026-10-15): Browserbase offers browser automation and agent infrastructure. Free tier: 1 browser hour ($5 in model tokens), 3 concurrent browsers, limited API calls. Developer $20/month: 100 browser hours then $0.12/hr overage, 25 concurrent browsers, usage-based overages on Search ($5/1k) and Fetch ($1-7/1k) calls. Startup $99/month: 500 browser hours then $0.10/hr, 100 concurrent browsers, higher limits with per-unit overages. Enterprise: custom pricing with 250+ browsers and usage-based billing. Model tokens billed separately on all plans. Free tier: yes. Interfaces: API Models: Any LLM selected by provider/model string (for example openai/gpt-5.4-mini) with your own API key Supports: byok Data: Gid Score 57/100, #34 overall; #4 in its category; 26,096.0 GitHub stars; usage +40% over 30 days; latest release 3.7.4 (2026-11-18).

## Price history (from archived copies of the official pricing page)
- Archived pricing pages read: 10 months, 2025-07 to 2026-04 (Internet Archive copies of https://www.browserbase.com/pricing)
- Cheapest paid plan: $20/mo in 2025-07, $20/mo in 2026-04; lowest seen $20/mo (2025-07)
- Price changes seen: 1: 2025-12: Scale plan no longer listed
- Full history with links to each archived copy: https://agentgid.com/pricing/history/stagehand/

## Alternatives in the same category
- Browser Harness: https://agentgid.com/agents/browser-harness/agent.md
- Browser Use: https://agentgid.com/agents/browser-use/agent.md
- agent-browser: https://agentgid.com/agents/agent-browser/agent.md
- ARTEMIS: https://agentgid.com/agents/google-artemis/agent.md
- Ego Lite: https://agentgid.com/agents/ego-lite/agent.md
- Open-AutoGLM: https://agentgid.com/agents/open-autoglm/agent.md

## Links
- AgentGid page: https://agentgid.com/agents/stagehand/
- Source: https://github.com/browserbase/stagehand
- Source: https://raw.githubusercontent.com/browserbase/stagehand/main/README.md
- Source: https://registry.npmjs.org/@browserbasehq%2Fstagehand/latest
- Source: https://pypi.org/pypi/stagehand/json
- Source: https://www.browserbase.com/pricing
- Source: https://discord.com/api/v10/invites/stagehand?with_counts=true

---
AgentGid's catalog is curated: every entry was checked against the maker's official pages (registry, repository, license and pricing pages) on the date shown; a license value shows the quote it stands on wherever AgentGid has one. Usage numbers come from AgentGid's own scan of public repositories set up for AI coding agents. Claims by makers are reported, not tested by AgentGid.
