Data Analysis

The data-analysis skill empowers AI agents to perform comprehensive statistical analysis, generate visualizations, and derive actionable insights from raw datasets. It supports various file formats including CSV, Excel, JSON, and SQL queries, enabling deep exploration of data structures and trends.

When to use it

• Analyzing large-scale sales or transaction logs to identify seasonal patterns and revenue anomalies.
• Processing user behavior metrics from web analytics tools to optimize conversion funnels.
• Generating dynamic charts and graphs to present complex findings to non-technical stakeholders.
• Cleaning and transforming messy raw data into structured formats ready for machine learning pipelines.

Key capabilities

• Multi-format ingestion: Reads CSV, Excel, JSON, Parquet, and SQL database connections.
• Statistical analysis: Calculates mean, median, standard deviation, correlation, and regression models.
• Visualization: Generates interactive plots including histograms, scatter plots, heatmaps, and time-series graphs.
• Data transformation: Performs filtering, grouping, aggregation, and pivoting on datasets.
• Report generation: Outputs summary statistics and visual reports directly within the chat interface.

Example prompts

• "Upload this CSV file of customer orders and calculate the average order value per region along with a bar chart."
• "Run a correlation analysis between website traffic and conversion rates, then visualize the top three correlated variables."
• "Load my SQL table sales_2023 and create a time-series line graph showing monthly revenue trends over the last year."

Tips & gotchas

Ensure your data files are properly formatted or accessible via a database connection before initiating analysis. For very large datasets, consider summarizing the data first to ensure optimal performance within the agent's processing limits.