---
id: agentql
version: "1.19.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# agentql — Tiny Fish AgentQL Python Client

License: permissive · Maintenance: active · Downloads: 74.2K/mo

## What it is and what it does

AgentQL is a Python client for Tiny Fish AgentQL, which replaces traditional CSS/XPath selectors with AI-driven natural language queries for web interaction. Instead of writing brittle element selectors, you describe what you want to extract or interact with in plain language, and the client handles locating and interacting with page elements through Playwright.

The package depends on Playwright for browser control, Pydantic for data validation, httpx and requests for HTTP operations, and includes CLI support via Typer. It's designed for developers building web automation, testing, or scraping workflows who want to avoid maintaining selector-based code that breaks when page layouts change.

Use it for:

- Automate web testing workflows without writing brittle CSS or XPath selectors.
- Extract structured data from websites using natural language descriptions instead of manual parsing.
- Build AI-powered bots that interact with web applications by understanding page content semantically.
- Prototype rapid web automation tasks without deep knowledge of page DOM structure.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

AgentQL is a Python client for interacting with web pages using AI-driven queries instead of traditional selectors, built on Playwright for browser automation.

Yes, if you're building web automation or scraping workflows and want to reduce selector maintenance overhead. The package is actively maintained, has no known vulnerabilities, low install friction, and uses a permissive license. Verify that AgentQL's query API and documentation match your use case before committing to it as a core dependency.

## Install

pip install agentql
uv add agentql
poetry add agentql

## Installing agentql

Before you install: Low friction: pure Python wheel with no compiled dependencies. Active maintenance (93 days since last release). Requires Python 3.9 or later.

License in practice: MIT license is permissive—you can use, modify, and distribute this package freely in commercial and private projects with minimal restrictions.

Quickstart:

pip install agentql

from agentql import AgentQL

agent = AgentQL()
result = agent.query("Find all product titles on this page")

Requires Python 3.9 or later and a working Playwright browser environment.

Verify before relying:

- Whether AgentQL's query language and API surface are documented sufficiently for typical use cases.
- Performance characteristics when handling large-scale scraping or complex page interactions.
- Whether the tf-playwright-stealth dependency affects compatibility with modern web applications.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 74.2K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

AI web scraping, playwright web automation, natural language web queries, browser automation testing, web interaction with AI, web-automation, ai-driven-queries, playwright-based

[View on SkillFed](https://skillfed.io/packages/agentql) · [View on PyPI](https://pypi.org/project/agentql/)