database-performance

Pass

Audited by Gen Agent Trust Hub on May 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill manages database credentials securely by referencing environment variables (${DB2i_HOST}, ${DB2i_USER}, ${DB2i_PASS}) rather than hardcoding sensitive information.
  • [SAFE]: All SQL tools are configured with readOnly: true and exclusively use SELECT statements, minimizing the risk of accidental or malicious data modification.
  • [SAFE]: The tool definitions employ parameterized SQL queries (e.g., using :schema_filter, :table_name), which effectively mitigates the risk of SQL injection from user-provided inputs.
  • [SAFE]: No evidence of prompt injection, remote code execution, or unauthorized data exfiltration was detected across any of the skill's files.
  • [SAFE]: The skill's behavior is consistent with its stated purpose of database performance monitoring.
Audit Metadata
Risk Level
SAFE
Analyzed
May 4, 2026, 09:52 PM
Security Audit — agent-trust-hub — database-performance