service-concierge-portal-generate
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of the
sf(Salesforce CLI) andcurltools to query, update, and deploy metadata to Salesforce orgs. These operations are performed via theBashtool and are essential for the skill's primary function of portal deployment and agent wiring. - [DYNAMIC_EXECUTION]: At runtime, the skill dynamically generates and modifies metadata files, including
PermissionSetandCspTrustedSiteXML records andDigitalExperienceBundleJSON files. It usespython3andsedwithin shell blocks to inject user-provided branding tokens and structural page modifications before deployment. - [INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface for indirect prompt injection as it ingests external data (user-provided site names and branding colors) via
AskUserQuestion. This data is subsequently interpolated into metadata files and shell commands. The skill includes basic sanitization instructions, such as alphanumeric filtering for URL paths, to manage this risk.
Audit Metadata