Kontext Data Agents help you analyze project data, generate charts, and create notebooks from natural language.
Current support on this page reflects the shipped team and plan session architecture.
Reference
| Capability | Support |
|---|---|
| Team architecture | Coordinator plus SQL, chart, notebook, and Python specialists |
| Data context | Uploaded files, connected catalogs, project lakehouse data |
| Knowledge context | Indexed project documents and curated runtime knowledge |
| Outputs | Answers, SQL, charts, notebooks |
| Session modes | team, plan |
| Runtime | User-configured LLM provider |
See AI Modes for mode behavior and LLM Providers for runtime support.
Available In
| Surface | Usage |
|---|---|
| Chat sessions | Ask questions, generate SQL, and work iteratively with project data |
| Team mode | Route execution across SQL, chart, notebook, and Python specialists |
| Plan mode | Ground a planning-first session in project and catalog context |
Example prompts
Suggest three questions based on project data
Suggest three questions based on the data in the project
Ask a project question
How many tables are in the data lake?
Generate SQL
Write a query to show transaction volume by year and month
Execute SQL
Execute the above query with a reasonable limit for previewing results
Ask a direct data question
Which age group has the highest number of customers?
Create a notebook from analysis
Create a notebook to visualise customer demographics by age band and country