experience-cms-content-type-generate
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the Salesforce CLI (sf) to perform project metadata operations, including sf project deploy for deployment and sf project retrieve for synchronization. It also performs targeted file removal of OS metadata files (.DS_Store) to ensure CLI command success.
- [INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect injection surface as it ingests and processes metadata descriptions from external Salesforce orgs via metadata-grounding and CLI retrieval. This data is used to populate discovery tables and influence schema generation. Ingestion points: Salesforce metadata retrieved from connected orgs. Boundary markers: Absent for retrieved metadata; results are displayed to the user for confirmation. Capability inventory: File system write access and metadata deployment via CLI. Sanitization: Implements an intent-sanity gate to validate user requests and enforces camelCase naming conventions for generated field APIs.
- [DATA_EXPOSURE]: The skill reads local project configuration files like sfdx-project.json and CMS schema files (schema.json) to perform its primary function of content type management and org synchronization.
Audit Metadata