home / mcp / playread mcp server
Enables Google search and content fetch as MCP tools via a Playwright-backed MCP server.
Configuration
View docs{
"mcpServers": {
"anentrypoint-playread": {
"command": "playread",
"args": [
"mcp"
]
}
}
}Playread is a Google search automation tool built on Playwright that can run as a command-line interface or as an MCP server. It exposes its flows as MCP tools so you can fetch page content or perform Google searches within your MCP-driven workflows. This makes it easy to add automated web search and content extraction to your agent pipelines without writing custom browser orchestration code.
You can run Playread as an MCP server to expose its flows as reusable tools in your MCP client. Start the server and then call the available tools from your client to perform actions like fetching page content or performing Google searches.
npm install -g playread
```
```
npx -y playread@latest google-searchTo run Playread as an MCP server, start it with the mcp flow. This exposes all flows as MCP tools that your client can discover and invoke.
playread mcp
```
```
// Example MCP client configuration
{
"mcpServers": {
"playread": {
"command": "npx",
"args": ["-y", "playread@latest", "mcp"]
}
}
}The server exposes two core tools you can use from your MCP client: fetch and google-search. fetch retrieves the main textual content from a URL and formats it in a structured way. google-search performs a Google search and returns the results with titles, URLs, and descriptions.
You can extend Playread by adding custom flows. A flow is a JavaScript module that exports an async function receiving a connected client instance. For example, you can navigate to a site, take a snapshot, or perform follow-up actions, and then close the browser when finished.
Performs a Google search and extracts results including titles, URLs, and descriptions.
Fetches a web page and extracts the main content into structured text with headings, paragraphs, lists, and code blocks.