experience-ui-bundle-site-generate
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves data from external sources and interpolates that data into generated metadata files without explicit sanitization.
- Ingestion points: The skill ingests data from
sfdx-project.json(namespace) and queries organization-level metadata (Organization.NamespacePrefix) and user-created records (UIBundle.DeveloperName) via the Salesforce CLI. - Boundary markers: The skill does not define clear boundaries or instructions for the agent to treat interpolated values as untrusted data during the templating process.
- Capability inventory: The skill possesses capabilities for reading/searching local files, writing generated metadata to the file system, and executing deployment validation via
sf project deploy. - Sanitization: There is no evidence of sanitization, escaping, or validation of the retrieved properties (such as
appDevNameorsiteName) before they are substituted into the XML/JSON templates. - [COMMAND_EXECUTION]: The skill utilizes the
sfCLI to perform data queries and project deployments. - Evidence: Instructions include the use of
sf data queryto resolve environment properties andsf project deploy validateto verify the generated metadata. - Context: These commands are standard operations for the Salesforce development lifecycle and align with the infrastructure management purpose of the skill as authored by the vendor.
Audit Metadata