r-explore

name	r-explore
description	Explore and inspect R data frames, objects, and environments interactively. Use when the user wants to understand data structure, preview data, or investigate R objects.

R Data Exploration

Use btw MCP tools for interactive R data exploration. These tools connect to a live R session with persistent state.

Key Tools

Tool	Purpose
btw_tool_env_describe_data_frame	Inspect structure, types, and summary of a data frame
btw_tool_env_describe_environment	List all objects in the R global environment
btw_tool_run_r	Run R code in the live session (state persists across calls)
btw_tool_sessioninfo_platform	Check R version, OS, locale
btw_tool_sessioninfo_is_package_installed	Check if a package is available

Workflow

1. Start with the environment — use btw_tool_env_describe_environment to see what's loaded
2. Inspect data frames — use btw_tool_env_describe_data_frame for structure and summary
3. Explore interactively — use btw_tool_run_r for custom queries, filtering, and aggregation

Common Exploration Patterns

# Preview rows
head(df, 10)
dplyr::glimpse(df)

# Value distributions
dplyr::count(df, column, sort = TRUE)
summary(df$column)

# Missing data
colSums(is.na(df))

# Unique values
purrr::map_int(df, dplyr::n_distinct)

# Cross-tabulation
table(df$col1, df$col2)

Anti-patterns

• Don't use Bash + Rscript for data exploration — use btw_tool_run_r to keep session state
• Don't call str() when btw_tool_env_describe_data_frame gives richer output
• Don't print() entire large data frames — use head() or dplyr::slice_sample()