skills/forcedotcom/sf-skills/service-itsm-agentic-setup-agentforce-studio-configure/Gen Agent Trust Hub
service-itsm-agentic-setup-agentforce-studio-configure
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
sf(Salesforce CLI) andnodeto execute local helper scripts (classify-enable-plan.mjs,record-enable-result.mjs,classify-final-report.mjs). These operations are used for managing and enabling Salesforce organization features. - [DATA_EXPOSURE]: The instructions explicitly emphasize security best practices by directing the agent to never extract or manually pass access tokens, instead relying on
sf api request restwhich handles authentication internally via the CLI's session. - [PROMPT_INJECTION]: The skill includes a required human-in-the-loop checkpoint (
AskUserQuestion) that forces the agent to present a plan and wait for explicit user confirmation before any state-mutating write operations are performed. - [DYNAMIC_EXECUTION]: The skill uses local Node.js scripts for deterministic decision-making (Standard A9). These scripts parse JSON data from local temporary files using
JSON.parse(), which is a safe practice for handling structured data. - [REMOTE_CODE_EXECUTION]: No remote script downloads or external package installations are performed; all logic is contained within the skill's distributed scripts and the trusted Salesforce CLI.
Audit Metadata