tables
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where row data is processed via Liquid templates, creating a risk that malicious data could influence system behavior.
- Ingestion points: Table row data is interpolated into
config.inputtemplates for action columns as defined inreferences/columns.md. - Boundary markers: The instructions do not define delimiters or explicit boundary markers to separate data from instructions during Liquid rendering.
- Capability inventory: The skill has the ability to trigger external integration actions via
api_writeon theaction-runsresource using the rendered templates. - Sanitization: No sanitization or validation logic is prescribed for row data before it is processed by the template engine.
- [DYNAMIC_EXECUTION]: The skill uses Liquid templates in action column configurations to dynamically generate input for integration actions at runtime based on the contents of table rows.
Audit Metadata