service-omni-permission-set-assign

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONPRIVILEGE_ESCALATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The script executes the Salesforce CLI (sf) and jq to manage organization data and assignments. These tools are used for their intended administrative purposes within the Salesforce ecosystem.
  • [DYNAMIC_EXECUTION]: The skill generates temporary XML manifests and project configuration files at runtime to deploy the Omni_Agent permission set. The generation logic uses a hardcoded whitelist of presence statuses to build the XML, preventing arbitrary content injection.
  • [PRIVILEGE_ESCALATION]: While the skill modifies user permissions, it includes a safety guard that blocks execution on production orgs by checking the IsSandbox, TrialExpirationDate, and OrganizationType fields. It also restricts assignments to users following a specific test-agent naming convention (agentN.suffix@example.com).
  • [PROMPT_INJECTION]: Input parameters such as the permission set names are validated against a strict regular expression (^[A-Za-z][A-Za-z0-9_]{0,79}$) before being interpolated into SOQL queries or shell commands, effectively preventing injection attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 01:11 PM
Security Audit — agent-trust-hub — service-omni-permission-set-assign