service-omni-presence-user-config-deploy

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests agent usernames and IDs from user input. Ingestion points: agent_usernames_csv argument in deploy-and-report.sh. Boundary markers: Absent. Capability inventory: sf data query and sf project deploy. Sanitization: The script uses strict regex validation and an XML entity escaping function to ensure that user-provided data cannot break the metadata structure or execute malicious commands.
  • [SAFE]: A 'safe_to_write' guard is implemented to check the target Salesforce organization's type (e.g., IsSandbox, OrganizationType) before executing any deployment, effectively preventing accidental changes to production customer environments.
  • [COMMAND_EXECUTION]: The skill executes standard Salesforce CLI (sf) commands to query org information and deploy metadata. All command arguments are derived from sanitized and validated user inputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:12 PM
Security Audit — agent-trust-hub — service-omni-presence-user-config-deploy