service-omni-agent-work-sharing-configure
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes the Salesforce CLI (
sf) andjqto query organization details and manage metadata. These operations are necessary for the skill's stated administrative purpose and are performed within the scope of the user's authenticated session. - [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external Salesforce environments via metadata retrieval and SOQL queries. The attack surface is effectively managed through the use of specific XML and JSON parsing rather than passing raw output directly to the LLM context, and all mutations require explicit user confirmation via CLI flags.
- [DYNAMIC_EXECUTION]: The test suite (
scripts/tests/_bootstrap.py) generates a temporary mock shell script to simulate the Salesforce CLI. This is a standard testing practice used to validate script behavior in a controlled environment and does not execute untrusted external code.
Audit Metadata