service-native-voice-recording-transcription-configure

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill performs administrative tasks within a Salesforce environment using the official Salesforce CLI (sf). It does not request or handle sensitive credentials, instead relying on the user's existing CLI authentication.
  • [COMMAND_EXECUTION]: The provided shell script executes standard utility commands (sed, grep, jq) and Salesforce CLI commands. These operations are constrained to a temporary working directory and are used solely for manipulating Metadata API XML files. The input parameters are validated (e.g., API version format check), and no vectors for arbitrary command injection were identified.
  • [DATA_EXFILTRATION]: The skill retrieves metadata configuration from the target org but does not transmit data to external third-party domains. It explicitly excludes session IDs, access tokens, and PII (like recording URLs or transcript content) from its output, focusing only on configuration status.
  • [EXTERNAL_DOWNLOADS]: No remote resources, scripts, or packages are fetched from the internet during execution. The skill's required tools (sf, jq) are declared as platform dependencies and are expected to be pre-installed.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 10:52 AM
Security Audit — agent-trust-hub — service-native-voice-recording-transcription-configure