The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes data_query and jupyter_execute tools to perform data transformations and visualizations. These tools execute Python and pandas code dynamically, which is consistent with the skill's primary purpose of data analysis and does not involve untrusted remote sources.
[PROMPT_INJECTION]: The skill processes external data files (CSV, JSON, etc.) from /workspace/data/. This represents an attack surface for Indirect Prompt Injection where instructions embedded in data files could attempt to influence the agent.
Ingestion points: External data files are loaded via the data_load tool from the /workspace/data/ directory.
Boundary markers: The instructions do not specify explicit delimiters or guardrails for handling untrusted content within data files.
Capability inventory: The agent has access to jupyter_execute and data_query for code execution, and data_save for writing files back to the workspace.
Sanitization: There are no mentioned mechanisms for sanitizing or validating the content of the data files before processing.

data-analysis