openai-api-development
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides patterns for interpolating untrusted user data into model prompts, creating a surface for indirect prompt injection. 1. Ingestion points: User-provided text in Chat Completions and image URLs/descriptions in Multimodal calls. 2. Boundary markers: The examples demonstrate standard API message structures but do not utilize specific delimiters or specialized framing to isolate untrusted content. 3. Capability inventory: The skill enables interactions with OpenAI models for text generation, vision analysis, and function calling. 4. Sanitization: The skill includes proactive security guidance, advising developers to validate and sanitize user inputs and use content moderation systems.
- [SAFE]: The skill promotes secure credential management by demonstrating the use of environment variables and explicitly warning against hardcoding or committing API keys.
- [SAFE]: The documented external dependencies are standard, well-maintained libraries for AI development and do not originate from untrusted or suspicious sources.
Audit Metadata