Kepler.gl AI Assistant Module
The AI Assistant is a module that adds an AI chatbot to Kepler.gl. This module aims to integrates Kepler.gl with AI-powered capabilities, enabling it to interact with multiple AI models seamlessly.
Overview
The system is designed to enable Kepler.gl, a React-based single-page application, to integrate an AI Assistant Module for performing tasks with large language models (LLMs) like OpenAI GPT models, Google Gemini models, Ollama models, etc.

Below is a flow map that shows how a user can update a basemap in Kepler.gl through a simple AI-driven prompt, showcasing the integration of LLMs with application actions and rendering.

The AI Assistant Module also provides a set of tools to support data analysis and visualization. These AI Tools are designed to be used in conjunction with the Kepler.gl application and transform kepler.gl into a powerful spatial data analysis and visualization tool. For more details about the AI Assistant Module, please to https://github.com/geodacenter/openassistant.
AI Tools
LLMs use these AI Tools to perform spatial data analysis and visualization tasks to help users explore and understand their data.
For example, a user can ask the AI Assistant to simply change the basemap to a voyager basemap, and the AI Assistant will call the basemap tool to change the basemap.
For complex tasks, the AI Assistant can use multiple tools to perform the task. For example, a user can ask the AI Assistant if the points dataset loaded in kepler.gl is clustering in zipcode areas. The AI Assistant could call the following tools to perform the task:
mapBoundary to get the boundary of current map view
queryUSZipcode to get a list of zipcodes using the map boundary
usZipcode to fetch the geometries of the zipcodes from Github site
saveData to save the zipcode areas as a new GeoJSON dataset in kepler.gl
spatialJoin to count the number of points in each zipcode area
saveData to save the spatialJoin result as a new dataset in kepler.gl
weightsCreation to create a e.g. queen contiguity weights using the spatialJoin result
local Moran's I to apply local Moran's I using the counts and the queen contiguity weights
saveData to save the local Moran's I result as a new dataset in kepler.gl
addLayer to add the local Moran's I result as a new layer in kepler.gl

These fine grained spatial tools are designed to transform the LLMs, which are fundarmentally statistical language models, into powerful spatial data analysis and visualization AI Agent.
As of May 2025, the AI Assistant Module supports the following AI Tools:
Tutorials
Next: Spatial Data Analysis using Kepler.gl AI Assistant