gpt-5-6-prompt-builder
Pass
Audited by Gen Agent Trust Hub on Jul 14, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references official documentation from OpenAI for prompting guidance. As this targets a well-known and trusted service for informational purposes, it does not pose a security risk.
- [SAFE]: No instances of obfuscation, hardcoded credentials, or malicious persistence mechanisms were detected. The skill operates entirely through natural language instructions and structured templates.
- [SAFE]: The skill instructions define an 'Operating contract' and 'Completeness gate' which act as logical guardrails to ensure the agent does not invent requirements or proceed without necessary user information.
- [PROMPT_INJECTION]: The skill processes user-provided inputs such as existing prompts, traces, and schemas to perform its transformation tasks. While this represents a data ingestion surface, the skill is designed for prompt engineering assistance and lacks access to sensitive capabilities that could be abused via indirect injection.
Audit Metadata