MCP Query Table is an MCP server that extracts and queries tabular financial data from online platforms using Playwright browser automation. It connects Model Context Protocol clients directly to financial web sources, including platforms such as iwencai, tdx, and eastmoney. Financial analysts, algorithmic traders, quantitative researchers, and software engineers use this server to automate the extraction of market figures without manually navigating dynamic browser pages. By driving headless browser sessions via Playwright, the server can render JavaScript-heavy web tables, handle interactive elements, and reliably extract up-to-date market metrics. It also facilitates multi-site data collection by providing fallback switching capabilities across supported sources. This enables AI assistants to retrieve structured stock screeners, price tables, and general corporate financial statistics to assist in equity research, portfolio tracking, and competitive intelligence directly within an LLM chat interface.
Category: Browser & Web Automation
Tags: financial-data, market data, playwright, web scraping
claude_desktop_config.json for Claude Desktop). 4. Add an entry for query-table under the mcpServers object with the appropriate runtime executable and path pointing to the cloned repository. 5. Restart your MCP client to enable the server tools.Part of MCP Servers
MCP Query Table is a Model Context Protocol server that uses Playwright automation to scrape and retrieve financial tables from web platforms such as iwencai, tdx, and eastmoney for AI clients.
The server controls browser sessions to extract structured financial tables from JavaScript-heavy portals, allowing AI assistants to query market data, screener metrics, and historical stock indicators directly with built-in source fallback.
It functions with standard Model Context Protocol clients that support local stdio server connections, such as Claude Desktop, Cursor, and custom agent frameworks configured with MCP tool-calling capabilities.
Specific package managers and commands are detailed in the repository. Users should clone the repository at https://github.com/MCP-Mirror/wukan1986_mcp_query_table, install the required environment dependencies and Playwright browsers, then configure their client JSON file.