fabriciq-ontology-consumption-cli

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFE
Full Analysis
  • Standard Tool Integration: The skill leverages common and verified command-line utilities such as az, jq, curl, and python3. These tools are used for legitimate administrative tasks like authentication, API interaction, and JSON processing.
  • Legitimate Data Processing: The skill includes routines to decode Base64 data returned by the Fabric 'Get Item Definition' API. This is a standard platform requirement for reconstructing item definitions and does not involve obfuscated malicious payloads.
  • Controlled Network Operations: All network requests target official and well-known Microsoft Fabric and Azure domains (e.g., api.fabric.microsoft.com, kusto.fabric.microsoft.com). These communications are performed using authenticated sessions established via az login.
  • Input Ingestion Surface: The skill processes ontology definitions which are external data sources. While this represents a theoretical surface for indirect prompt injection—where an attacker could attempt to influence agent behavior through a malicious schema—the risk is mitigated by the skill's read-only nature and its specific focus on mapping logical properties to physical data columns.
  • Well-Defined Consumption Scope: The skill explicitly restricts itself to read-only operations and delegates data manipulation or authoring tasks to other specialized skills, adhering to the principle of least privilege for the current session.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 02:44 PM
Security Audit — agent-trust-hub — fabriciq-ontology-consumption-cli