What are the prerequisites?
Node.js v20+, npm/yarn, an OpenAI API key (for GPT models), and a Google AI key (for Gemini models).
@tmahesh
Basic information
agentkit for playwright-mcp server
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"playwright-agent": {
"command": "npx",
"args": [
"@playwright/mcp@latest",
"--port",
"8931"
]
}
}
}Tool list not available yet
This listing does not currently have a tool list to display. This does not mean the server has no tools; check its documentation for supported capabilities.
Fetch the live tool list by running this server in a temporary sandbox using the button above.
AgentKit Browser Automation is a browser agent built with Inngest-AgentKit and Firebase-Genkit, featuring an agent system for intelligent web navigation and task execution. It integrates with Playwright MCP to control a browser via an MCP server.
Install Node.js v20+, clone the repository, install dependencies, and set environment variables for OpenAI and/or Google AI keys. Start the Playwright MCP server on port 8931, then run either npx tsx genkit.ts (for Genkit SDK) or npx tsx agentkit.ts (for Inngest-AgentKit). Optional dev servers provide built‑in observability.
Node.js v20+, npm/yarn, an OpenAI API key (for GPT models), and a Google AI key (for Gemini models).
Run `npx @playwright/mcp@latest --port 8931` in a separate terminal window before launching the agent.
The Genkit mode (`npx tsx genkit.ts`) uses the Genkit SDK, while the AgentKit mode (`npx tsx agentkit.ts`) uses Inngest-AgentKit. Both achieve browser automation but differ in their agent framework.
Yes. For Genkit mode, run `npx genkit start -- npx tsx --watch genkit.ts`. For AgentKit mode, run `npx inngest-cli@latest dev --no-discovery -u http://localhost:3000/api/inngest -v`. Both provide built‑in OpenTelemetry observability.
A flexible HTTP fetching Model Context Protocol server.
Fast, lightweight Firecrawl/Tavily alternative in Rust. Web scraper, crawler & search API with MCP server for AI agents. Drop-in Firecrawl-compatible API (/scrape, /crawl, /search). 2.3x faster than Tavily, 1.5x faster than Firecrawl in 1K-URL benchmarks. 6 MB RAM, single binary.
Capture screenshots of webpages as images or PDFs with Screenshot Scout.
Helps AI assistants access text content from bot-protected websites. MCP server that fetches HTML/markdown from sites with anti-automation measures using Scrapling.
Comments