The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external warehouse tables and views via system.information_schema and posthog:execute-sql. This creates a surface where data processed by the agent could contain instructions, though this is inherent to the skill's purpose as a data validation tool.
Ingestion points: Data is pulled from system.information_schema.data_quality_checks, system.information_schema.columns, and the output of posthog:execute-sql in SKILL.md.
Boundary markers: The instructions do not define explicit boundary markers or delimiters for untrusted data content.
Capability inventory: The skill utilizes posthog:execute-sql for reading/querying and posthog:data-quality-check-create-* for writing check configurations.
Sanitization: No specific sanitization or escaping logic is described for the content retrieved from warehouse tables.
[DYNAMIC_EXECUTION]: The skill enables the generation and execution of dynamic HogQL queries through the custom_sql check type and the manual execution of compiled queries via posthog:execute-sql. This is the primary intended function of the skill for performing arbitrary data assertions.

authoring-data-quality-checks