openai-agents-python
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns or security risks were identified in the skill instructions or metadata. The content is strictly educational and design-oriented.
- [EXTERNAL_DOWNLOADS]: The skill references the official OpenAI Agents Python repository on GitHub. This is a reference to a well-known service and organization for development purposes.
- [INDIRECT_PROMPT_INJECTION]: While the skill provides instructions for building agents that handle user input, it proactively addresses security by recommending the implementation of guardrails, prohibited action policies (such as blocking destructive commands or external data transfers), and the use of fixed JSON schemas for output validation.
Audit Metadata