skills/aliyun/maxcompute-semantic/udf/Gen Agent Trust Hub

udf

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates legitimate management of cloud resources (MaxCompute UDFs) using the vendor-provided CLI tool mcs.
  • [SAFE]: Instructions explicitly require user confirmation for any destructive or state-changing operations (creating or removing UDFs and resources), which is a recommended safeguard against autonomous agent errors.
  • [SAFE]: The testing functionality for UDFs includes detailed constraints on input arguments, specifically limiting them to SQL literals and prohibiting expressions or identifiers to prevent SQL injection vulnerabilities.
  • [SAFE]: No unauthorized data access, external network exfiltration, or obfuscated instructions were found in the provided files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 02:28 PM
Security Audit — agent-trust-hub — udf