services-extension-consumption

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides documentation for TerminalService.simpleExec, which allows executing shell commands. This is used for integration with the Salesforce CLI (sf) and includes built-in handling for environment variables like SF_LOG_LEVEL and SF_DISABLE_TELEMETRY.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: Documents access to the ConnectionService and SettingsService, which manage Salesforce authentication tokens and instance URLs. These are standard operations for Salesforce development tools to facilitate secure communication with the platform.
  • [DYNAMIC_EXECUTION]: Mentions FsService.readJSON which uses Effect.Schema for validated deserialization. This is a secure practice to prevent issues with corrupted or malicious JSON data from external sources.
  • [INDIRECT_PROMPT_INJECTION]: The skill describes methods for ingesting workspace data such as sfdx-project.json and metadata manifests. While this represents a data ingestion surface, the documentation encourages the use of structured schemas and standard Salesforce APIs, which include their own validation layers.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 10:06 AM
Security Audit — agent-trust-hub — services-extension-consumption