openai-platform-ops

Pass

Audited by Gen Agent Trust Hub on Aug 8, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an educational and operational reference for OpenAI platform administration, containing no executable malicious payloads.
  • [SAFE]: Instructions for handling sensitive data (API keys) consistently recommend secure practices like environment variables, secret managers, and file permission hardening (e.g., umask 077).
  • [SAFE]: External references and dependencies are limited to official OpenAI APIs, the official Terraform provider, and well-known libraries such as the Python openai SDK, requests, and the oci SDK.
  • [SAFE]: The skill provides defensive guidance, correctly identifying and offering mitigations for prompt injection and data leakage risks within agent-builder workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 8, 2026, 03:02 AM
Security Audit — agent-trust-hub — openai-platform-ops