authoring-data-quality-checks
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted metadata from database tables such as
system.information_schema.columnsanddata_quality_checks. This data enters the agent's context without explicit boundary markers or sanitization, creating a surface where malicious metadata could influence agent behavior. The skill possesses extensive capabilities including executing arbitrary HogQL and creating persistent data quality checks. - [DYNAMIC_EXECUTION]: The skill facilitates the creation and execution of
custom_sqlassertions, which are dynamically assembled HogQL queries run at runtime to validate table invariants. - [COMMAND_EXECUTION]: The skill uses specialized tools like
posthog:execute-sqlto interact with the data warehouse and run complex queries based on its analysis of the schema.
Audit Metadata