The Agent Skills Directory

[PROMPT_INJECTION]: The skill implements a data quality reporting mechanism that incorporates external data into the agent's context, creating a potential indirect prompt injection surface.
Ingestion points: Data is ingested from external data assets (tables) via the Great Expectations library, as seen in the DataQualityPipeline class in quality_pipeline.py.
Boundary markers: The generate_report function does not use clear delimiters or instructions to treat the observed_value as untrusted content when presenting it to the agent/user.
Capability inventory: The skill utilizes the great_expectations library to perform network-based or file-based data reads from configured data sources.
Sanitization: The logic in quality_pipeline.py directly interpolates observed_value from validation results into a Markdown report without sanitization or escaping.

data-quality-frameworks