looker-performance-optimizer

Pass

Audited by Gen Agent Trust Hub on Jul 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses official vendor-provided tools (Looker CLI and Looker MCP) to interact with Looker projects and database connections. These interactions are consistent with the skill's primary purpose of performance optimization.
  • [SAFE]: No obfuscation, data exfiltration, or hardcoded credentials were found in the skill files. All references to database connections and ETL metadata use generic placeholders or standard administrative table names.
  • [SAFE]: The documentation includes example bash scripts for cache-warming workflows. These are provided as educational templates for developers and do not represent unauthorized or hidden command execution by the agent.
  • [SAFE]: The skill defines a clear, multi-step workflow for query diagnostics that includes using subagents for parallel processing. This is an architectural optimization for agentic execution and does not bypass safety constraints.
  • [SAFE]: Although the skill processes external data such as SQL execution plans and LookML files, it does so within a controlled developer environment to produce performance reports, and no indirect prompt injection vulnerabilities were identified in the logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 27, 2026, 12:06 PM
Security Audit — agent-trust-hub — looker-performance-optimizer