omnistudio-dependencies-analyze
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill identifies an attack surface where it ingests and processes metadata from Salesforce records.
- Ingestion points: Metadata fields such as
PropertySetConfig,DataSourceConfig, andDefinitionare retrieved from Salesforce objects using thesfCLI. - Boundary markers: The instructions do not explicitly require delimiters or warnings to disregard instructions that might be embedded in the retrieved JSON metadata.
- Capability inventory: The skill uses
sf data queryfor metadata retrieval andsf api request restfor interaction with Salesforce APIs. - Sanitization: Technical dependencies are extracted by parsing JSON structures without specific sanitization of the string content for embedded prompt instructions.
Audit Metadata