The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses sf api request rest commands via Bash to interact with Salesforce APIs for reading and enabling feature toggles. All commands are structured according to standard Salesforce CLI practices.
[COMMAND_EXECUTION]: Node.js scripts are used as deterministic logic engines to process API responses and compute enablement plans. These scripts are invoked via Bash using the node CLI.
[REMOTE_CODE_EXECUTION]: No remote code execution was detected. All scripts are local to the skill directory, and the skill explicitly forbids manual token extraction or external curl calls.
[DATA_EXFILTRATION]: No data exfiltration detected. Network operations are limited to the official Salesforce API via the sf CLI, and sensitive information like access tokens are never handled directly by the skill instructions or scripts.
[CREDENTIALS_UNSAFE]: No hardcoded credentials found. The skill leverages the existing authenticated session in the Salesforce CLI (sf), maintaining a secure credential management posture.
[PRIVILEGE_ESCALATION]: The skill operates within the context of the user's authenticated Salesforce session. It does not attempt to acquire elevated system-level permissions (e.g., sudo) or bypass org-level permission checks.
[INDIRECT_PROMPT_INJECTION]: While the skill ingests data from external API responses, it processes this data using deterministic Node.js scripts with fixed classification tables rather than direct prompt interpolation, significantly mitigating the risk of indirect injection.

service-itsm-agentic-setup-agentforce-studio-configure