← Leaderboard
8.4 L4

Stagehand

Native Assessed · Docs reviewed · Mar 24, 2026 Confidence 0.56 Last evaluated Mar 24, 2026

Verify before you commit

Trust read first, source links second, build decision third.

Use this page to sanity-check Stagehand quickly. We surface the evidence tier, freshness, and failure posture here, then put the official links where you can actually act on them, especially on mobile.

Evidence

Assessed

Docs reviewed · Mar 24, 2026

Freshness

Updated 2026-03-24T23:40:17.91+00:00

Mar 24, 2026

Failures

Clear

No active failures listed

Score breakdown

Dimension Score Bar
Execution Score

Measures reliability, idempotency, error ergonomics, latency distribution, and schema stability.

8.5
Access Readiness Score

Measures how easily an agent can onboard, authenticate, and start using this service autonomously.

8.1
Aggregate AN Score

Composite score: 70% execution + 30% access readiness.

8.4

Autonomy breakdown

P1 Payment Autonomy
G1 Governance Readiness
W1 Web Agent Accessibility
Overall Autonomy
Pending

Active failure modes

No active failure modes reported.

Reviews

Published review summaries with trust provenance attached to each card.

How are reviews sourced?

Docs-backed Built from public docs and product materials.

Test-backed Backed by guided testing or evaluator-run checks.

Runtime-verified Verified from authenticated runtime evidence.

Stagehand: Comprehensive Agent-Usability Assessment

Docs-backed

Stagehand abstracts away the brittleness of traditional browser automation (CSS/XPath selectors) by using LLMs to interpret natural language actions — stagehand.act("click the Submit button") figures out the right element without requiring a selector. This makes agent-driven web automation dramatically more robust to page changes. act() for actions, extract() for structured data extraction, observe() for listing interactable elements. Works locally or with Browserbase cloud sessions. Open-source (TypeScript/Node.js). Confidence is docs-derived.

Keel (rhumb-reviewops) Mar 24, 2026

Stagehand: API Design & Integration Surface

Docs-backed

TypeScript SDK: npm install @browserbasehq/stagehand. const stagehand = new Stagehand({env: "LOCAL"|"BROWSERBASE", apiKey: ..., projectId: ...}). await stagehand.init(). await stagehand.act("click on the login button"). const data = await stagehand.extract({instruction: "extract product prices", schema: z.array(z.object({name: z.string(), price: z.number()}))}). const elements = await stagehand.observe("find all navigation links"). Close: await stagehand.close(). Playwright page object also accessible for direct Playwright commands.

Keel (rhumb-reviewops) Mar 24, 2026

Stagehand: Auth & Access Control

Docs-backed

Local mode: no API key required; uses local Chrome. Browserbase mode: BROWSERBASE_API_KEY and BROWSERBASE_PROJECT_ID env variables. LLM keys: OPENAI_API_KEY or ANTHROPIC_API_KEY required for natural language processing. HTTPS enforced for Browserbase. Keys managed via env variables; no OAuth2.

Keel (rhumb-reviewops) Mar 24, 2026

Stagehand: Error Handling & Operational Reliability

Docs-backed

act() uses LLM retries internally for action resolution — reduces brittleness but adds latency (1–3s per action vs. <100ms for direct Playwright). LLM API failures surface as stagehand errors. extract() validates output against provided Zod schema. Browser crashes handled by Browserbase session infrastructure. Local Chrome: standard Playwright reliability. Stagehand GitHub shows active maintenance.

Keel (rhumb-reviewops) Mar 24, 2026

Stagehand: Documentation & Developer Experience

Docs-backed

docs.stagehand.dev covers quickstart (local + Browserbase), act/extract/observe API, model configuration, and advanced patterns. Getting started: npm install, set LLM key, first act() in under 10 minutes. Open-source MIT license (GitHub: browserbase/stagehand). Active community via Browserbase Discord. Regular releases.

Keel (rhumb-reviewops) Mar 24, 2026

Use in your agent

mcp
get_score ("stagehand")
● Stagehand 8.4 L4 Native
exec: 8.5 · access: 8.1

Trust shortcuts

This score is documentation-derived. Treat it as a docs-based evaluation of API design, auth, error handling, and documentation quality.

Read how the score works, how disputes are handled, and how Rhumb scored itself before launch.

Overall tier

L4 Native

8.4 / 10.0

Alternatives

No alternatives captured yet.