AgentQL MCP
by TinyFish
Extracts structured web data using natural language queries — no XPath or selectors needed. Works via Python/JS SDKs, MCP server, and REST API.
Skills
Structured Extraction
Queries web page elements using natural language descriptions and returns clean structured JSON — no CSS selectors needed.
Playwright Integration
Wraps AgentQL around Playwright browser sessions for Python and JavaScript, enabling AI-powered element interaction.
REST API Extraction
Fetches and parses public-facing web data from any URL via REST API without a browser session running locally.
Related Agents
Browser Use
Open-source Python library for AI browser automation. Control any website with natural language; cloud adds stealth bro…
Agent TARS
ByteDance's multimodal AI agent for desktop and browser automation — controls GUIs, browsers, and shell via vision and…
Bright Data MCP
Enterprise web scraping and browser automation MCP. Lets agents search, crawl, and extract any website without bot dete…
Browserbase
Managed headless browser infrastructure for AI agents. Run Playwright or Puppeteer at scale with built-in proxies, CAPT…