audit-permissions

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • Dataverse Metadata Discovery: The skill executes specialized Node.js scripts to fetch lookup and relationship metadata from the Dataverse API. This ensures the permission audit is grounded in the actual environment schema and identifies potential 403 Forbidden scenarios.\n- Local Tool Execution: It uses the Bash tool and Power Platform CLI (pac) to discover project context and validate local configuration files. These actions are standard for development and auditing tasks within the Power Platform ecosystem.\n- Indirect Data Ingestion: The skill analyzes local site source code and configuration files to build a map of table usage. While this involves processing untrusted local content, the data is used strictly for report generation and logic verification.\n- Automated Report Generation: Findings are compiled into a visual HTML report using a rendering script. This process is isolated and does not modify existing site source code without explicit user direction.\n- Remediation Orchestration: If gaps are identified, the skill can invoke a secondary agent to propose fixes. This modular approach ensures that analysis and modification are handled as separate, user-approved steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 11:50 PM