Mode d’emploi
- Copiez la commande et examinez le paquet avant de l’installer dans votre environnement agent.
- Consultez le document original pour les outils, identifiants et environnements requis.
- Décrivez les entrées et le livrable souhaité ; comparez la sortie aux sources.
npx skills add https://github.com/smartworldarafath/rea_skill --skill hyperbrowser-mcpAvant de commencer
- L’installation n’inclut ni abonnement au modèle, ni quota API, ni accès aux fichiers.
- Étoiles et installations reflètent l’activité, pas un test indépendant de qualité ou de sécurité.
Rendu depuis le dépôt source en conservant titres, exemples, code, tableaux, liens et images.
Hyperbrowser MCP (Cloud Headless Browser Automation & Crawling)
Based on hyperbrowserai/mcp: an MCP server providing cloud headless browser automation, dynamic JavaScript rendering, full-site crawling, and structured web data extraction for AI agents.
🎯 When to Use
- Complex Single-Page Applications (SPAs): Scraping websites that rely on heavy client-side JavaScript, React/Vue hydration, or WebGL where static HTML extractors return blank templates.
- Deep Site Crawling: Recursively following links across web applications, documentation trees, or API portals.
- Stealth Browsing & Session Recordings: Recording video replays and interaction traces of web navigation for verification and audit trails.
- Browser Agent Delegation: Powering browser agents (such as Claude Computer Use, OpenAI CUA, or Browser Use).
🚀 Setup & Registration
MCP Configuration
Register Hyperbrowser MCP in your agent's configuration (.mcp.json or IDE settings):
{
"mcpServers": {
"hyperbrowser": {
"command": "npx",
"args": ["-y", "hyperbrowser-mcp"],
"env": {
"HYPERBROWSER_API_KEY": "YOUR_HYPERBROWSER_API_KEY"
}
}
}
}🛠️ Key Capabilities & Tools
crawl_page: Crawls dynamic web applications and returns fully rendered DOM contents and structured link trees.extract_data: Uses vision and LLM parsing to turn visual elements on web pages into clean JSON data models.record_session: Records full interactive browser sessions into shareable artifacts for test reports and audits.
⚡ Integration with REA & Security Audits
- Auditing Client-Side Rendered Applications: Render Next.js, Nuxt, or Vite web apps in Hyperbrowser to intercept WebSocket connections, runtime-injected scripts, and dynamic API calls that do not appear in raw HTML.
- Visual Regression Verification: Capture live renders of web applications after applying security headers or CSP rules to confirm user interface integrity.