The Agent Skills Directory

[SAFE]: The skill implements a professional and standard data science workflow, utilizing well-known R packages such as tidyverse, fixest, and modelsummary. It emphasizes best practices like setting random seeds and organizing outputs into dedicated directories.
[COMMAND_EXECUTION]: The skill is authorized to use the Bash tool to execute R scripts and manage the project's file structure. This capability is necessary for the skill's primary function of automated data analysis and is applied in a structured manner.
[PROMPT_INJECTION]: The skill takes user input via $ARGUMENTS to define the analysis goal and data sources, which creates a surface for indirect prompt injection.
Ingestion points: Dataset paths and natural language descriptions of analysis goals in the SKILL.md file.
Boundary markers: None are explicitly used to delimit user input from the rest of the prompt logic.
Capability inventory: The skill has access to Bash, Write, Edit, and Task tools, which allow for code execution and file system modifications.
Sanitization: The skill mitigates risks by requiring the delegation of generated scripts to an r-reviewer agent for validation before final execution and presentation of results.

data-analysis