stagehand-expert
🎭 Skill: Stagehand Expert (v4.1.0)
Executive Summary
The stagehand-expert is the elite specialist in browser automation and high-precision agent orchestration. In 2026, web automation has shifted from brittle selectors to Natural Language Primitives and Direct CDP Communication. This skill focuses on mastering Stagehand V3, leveraging Decision Caching for zero-cost CI/CD, and navigating complex Shadow DOM/iframe structures with 44% more velocity.
📋 Table of Contents
- Proactive Investigation Protocol
- The "Do Not" List (Anti-Patterns)
- Core Primitives (Act, Extract, Observe)
- Direct CDP & Performance
- Advanced Agent Caching
- Autonomous Agents (CUA)
- Reference Library
🔍 Proactive Investigation Protocol
Before writing a single test, the expert MUST perform a Deep Discovery:
- Route Mapping: identify the user flow from
page.tsxor router configs. - UI Component Audit: Read source code to find IDs, labels, and loading states.
- Vibe Check: Measure layout stability using the CDP "Vibe Score."
- Schema Inference: Analyze existing backend/DB types to create 100% compatible
extract()Zod schemas.
🚫 The "Do Not" List (Anti-Patterns)
| Anti-Pattern | Why it fails in 2026 | Modern Alternative |
|---|---|---|
| Manual Frame Switching | Fragile and slow. | Use DeepLocator (>>) & CDP. |
| Hardcoded Wait(2000) | Unreliable and causes jank. | Use domSettleTimeout. |
| Missing finally { close() } | Leaves zombie processes. | Mandatory try...finally. |
| LLM Calls in CI | Slow and expensive. | Use Persistent Decision Caches. |
| Ignoring CSS Animations | Interactions fail during transitions. | Use Reanimated-aware Waiters. |
⚡ Core Primitives Mastery
- Act: Precise natural language instructions with mapped variables.
- Observe: Single-turn identification of all page elements for 70% cost reduction.
- Extract: Structured, Zod-validated data pulling with semantic flattening.
💾 Advanced Decision Caching
Transform E2E tests into a deterministic asset:
- Develop Locally: Live LLM generates the cache.
- Commit Cache: Store DOM snapshots and results in Git.
- Zero-Cost CI: Run tests in "Cached-Only" mode.
See References: Agent Caching for details.
🤖 Autonomous Agents & CUA
For the most complex UIs (Cross-origin iframes, dynamic canvas):
- Computer Use Agent (CUA): Pure visual reasoning for impossible-to-parse elements.
- Safety Callbacks: Mandatory human-in-the-loop for financial or destructive actions.
📖 Reference Library
Detailed deep-dives into Stagehand Excellence:
- Direct CDP Communication: Velocity and deep access.
- Agent Caching: Determinism and cost savings.
- Shadow DOM Mastery: Jumping документ boundaries.
- Installation & Setup: The Bun/Playwright stack.
Updated: January 22, 2026 - 21:20
More from yuniorglez/gemini-elite-core
filament-pro
Master of Filament v4 (2026), specialized in Custom Data Sources, Nested Resources, and AI-Augmented Admin Panels.
80remotion-expert
Senior Specialist in Remotion v4.0+, React 19, and Next.js 16. Expert in programmatic video generation, sub-frame animation precision, and AI-driven video workflows for 2026.
58tailwind4-expert
Senior expert in Tailwind CSS 4.0+, CSS-First architecture, and modern Design Systems. Use when configuring themes, migrating from v3, or implementing native container queries.
48pdf-pro
Master of PDF engineering, specialized in AI-driven extraction, high-fidelity Generation (Puppeteer), and PDF 2.0 Security.
46ui-ux-specialist
Senior Accessibility & Frontend Engineer. Expert in WCAG 2.2 standards, Semantic HTML, and Inclusive Design for 2026.
37threejs-expert
Senior WebGPU & 3D Graphics Architect for 2026. Specialized in Three.js v172+, WebGPU-first rendering, TSL (Three Shader Language), and high-performance React 19 integration via `@react-three/fiber` and `@react-three/drei`. Expert in building immersive, low-latency, and accessible 3D experiences for the modern web.
36