data-import
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the
SN-Execute-Background-Scripttool for administrative tasks such as retrying failed import rows, monitoring throughput, and rolling back changes. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface through JDBC, LDAP, and REST data sources (Phase 1). Boundary markers for external data are absent. The capability inventory includes server-side script execution and table modifications. Sanitization is addressed in Step 5.3, which demonstrates using
onBeforetransform scripts for regex validation, null checking, and string normalization. - [COMMAND_EXECUTION]: The skill uses
SN-Batch-CreateandSN-Batch-Updateto manage dictionary entries and transform mappings in bulk.
Audit Metadata