AgentQL

AgentQLTools enable an Agent to browse and scrape websites using the AgentQL API.

Prerequisites

The following example requires the agentql library and an API token which can be obtained from AgentQL.

1uv pip install -U agentql

1export AGENTQL_API_KEY=***

Example

The following agent will open a web browser and scrape all the text from the page.

1from kern.agent import Agent
2from kern.models.openai import OpenAIResponses
3from kern.tools.agentql import AgentQLTools
4
5agent = Agent(
6    model=OpenAIResponses(id="gpt-5.2"), tools=[AgentQLTools()]
7)
8
9agent.print_response("https://kern.ndx.rocks/introduction", markdown=True)

Note

AgentQL will open up a browser instance (don't close it) and do scraping on the site.

Toolkit Params

Parameter	Type	Default	Description
`api_key`	`str`	`None`	API key for AgentQL
`scrape`	`bool`	`True`	Whether to use the scrape text tool
`agentql_query`	`str`	`None`	Custom AgentQL query
`enable_scrape_website`	`bool`	`True`	Enable the scrape_website functionality.
`enable_custom_scrape_website`	`bool`	`True`	Enable the custom_scrape_website functionality.
`all`	`bool`	`False`	Enable all functionality.

Toolkit Functions

Function	Description
`scrape_website`	Used to scrape all text from a web page
`custom_scrape_website`	Uses the custom `agentql_query` to scrape a web page

Developer Resources

View Tools