bx-ai-chatting
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill serves as a technical reference for BoxLang's AI integration features. All code samples are educational and follow standard developer documentation practices.
- [CREDENTIALS_UNSAFE]: The documentation mentions API key configuration and includes placeholders such as
sk-...andsk-ant-.... These are standard, non-functional placeholders used for demonstration and do not represent a security risk. - [INDIRECT_PROMPT_INJECTION]: The skill documents functions (
aiChat,aiChatAsync,aiChatStream) that process message strings for LLM interaction. While this creates a surface for indirect prompt injection if untrusted data is passed to these functions, this is an inherent property of AI client libraries rather than a malicious pattern in the skill. - Ingestion points: The
messageparameter inaiChat(),aiChatAsync(), andaiChatStream()(found in SKILL.md). - Boundary markers: Not present in the basic usage examples.
- Capability inventory: Network operations to AI providers (OpenAI, Claude, etc.) via the documented BIFs.
- Sanitization: Not present in the basic usage examples.
Audit Metadata