platform-agentexchange-partner-offers-configure

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses sf (Salesforce CLI) and jq commands to interact with the Salesforce platform and project configuration. These are standard tools for Salesforce developers.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads the local sfdx-project.json configuration file and queries org preferences via the Salesforce Tooling API to determine current status. No sensitive credentials or unauthorized network exfiltration was detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local project files and user-supplied org aliases.
  • Ingestion points: Reads the project directory path from sfdx-project.json and org aliases from user input.
  • Boundary markers: The instructions mandate human-in-the-loop confirmation of the target org and state before executing deployment commands.
  • Capability inventory: Includes shell command execution (query/deploy) and local file writes (XML metadata).
  • Sanitization: Relies on the agent's internal handling of variable interpolation into shell templates.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 12:31 PM
Security Audit — agent-trust-hub — platform-agentexchange-partner-offers-configure