The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the run_python_snippet tool to execute dynamically generated Python code for data processing and analysis.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by reading external data files and incorporating that information into code generation tasks.
Ingestion points: The skill uses read_files and pd.read_csv to ingest data from CSV and Excel files.
Boundary markers: There are no specific instructions or delimiters provided to ensure the agent ignores instructions embedded within the data.
Capability inventory: The agent has access to the run_python_snippet tool to execute code.
Sanitization: The instructions do not include steps to sanitize or validate the content of the data files before they are processed.

data-qa