Search 1,000 Pages in Seconds

Keenable SELECT is a revolutionary agent that enables users to search the web in SQL, allowing for precise and efficient extraction of specific information from live web data. This innovative tool uses a combination of web search, semantic extraction, and DuckDB to provide accurate results.

The SELECT tool is capable of searching more than 1,000 pages and filtering them with an exact WHERE clause at no LLM cost, making it a powerful solution for data extraction. Additionally, the tool can extract fields with one small LLM call per row and group the rows, streamlining the data collection process.

Keenable SELECT has a server that saves every query result as a result set with an id, allowing for easy reference and reuse of previous queries. The generate_html_report function takes a brief and result set ids and returns a shareable link to an HTML report, making it simple to share and collaborate on research findings.

The research agent is a plain tool loop, consisting of an LLM with the select tool, which writes and runs its own queries until it can answer, and streams its tool calls, results, and answer as events. This allows for a transparent and reproducible research process.

## What makes Keenable SELECT unique Keenable SELECT uses WEB_SEARCH and WEB_FETCH operators that can run per row and use row columns as arguments, enabling precise and targeted data extraction. The server finds the operators in the parsed statement, runs them, and replaces them with plain columns, allowing for seamless integration with traditional SQL queries.

## Why it matters Keenable SELECT has the potential to revolutionize the way we conduct research and extract data from the web. By providing a fast, accurate, and efficient way to search and extract specific information, this tool can save time and resources for researchers, businesses, and individuals alike. With its ability to search large volumes of data and filter results with precision, Keenable SELECT is an invaluable resource for anyone looking to extract insights from the web.

The implications of Keenable SELECT are far-reaching, and its applications can be seen in various fields, from academic research to business intelligence. As the amount of data available on the web continues to grow, tools like Keenable SELECT will become increasingly important for making sense of this data and extracting valuable insights.