transform-maps
Warn
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [DYNAMIC_EXECUTION]: The skill generates and implements JavaScript (ES5) logic for ServiceNow transform scripts such as onBefore, onAfter, and onComplete, as well as scripted field mappings. These scripts are stored as strings in database fields (e.g., source_script) and executed at runtime by the ServiceNow transformation engine. Additionally, the skill documentation references the use of the snow_execute_script tool for running arbitrary scripts.
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion and processing of data from untrusted external sources (JDBC databases, REST APIs, and CSV files).
- Ingestion points: Data enters the environment via sys_data_source configurations and is staged in import set tables (e.g., u_employee_import) before processing.
- Boundary markers: The provided examples lack specific boundary delimiters or instructions to ignore embedded commands within the processed data streams.
- Capability inventory: The skill leverages snow_execute_script, snow_create_transform_map, and GlideRecord operations to perform database updates, inserts, and logic execution based on the imported data.
- Sanitization: While basic field validation (null checks) and simple normalization (e.g., converting email to lowercase) are provided in templates, there is no comprehensive sanitization logic to prevent malicious payloads in the source data from influencing the transform script execution flow.
Audit Metadata