The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes validation results from external data sources and includes specific data values in its output reports and notifications.
Ingestion points: The generate_report method in quality_pipeline.py and the Slack notification action in orders_checkpoint.yml consume observed_value data from dataset validation results.
Boundary markers: Absent; external data values are interpolated directly into the markdown report and Slack message strings.
Capability inventory: The skill generates reports for the agent's context and can send external notifications via Slack.
Sanitization: Absent; values retrieved from the data source are not escaped or validated before being included in the reports.
[EXTERNAL_DOWNLOADS]: Fetches and installs the great_expectations package, which is a standard, well-known framework for data quality validation.

data-quality-frameworks