ai-orchestration-llamaindex
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides explicit guidance on secure credential management, instructing users to use environment variables and the dotenv package instead of hardcoding API keys.
- [SAFE]: All tool definitions in the examples use Zod schemas for parameter validation, which ensures that agent inputs are properly structured and typed before execution.
- [SAFE]: The skill includes critical safety reminders, specifically warning against the use of eval() in custom tools and recommending safe math parsers instead.
- [SAFE]: Dependencies are restricted to official LlamaIndex.TS provider packages and well-known, reputable libraries like Zod and Tree-sitter.
- [SAFE]: The document loading patterns (e.g., SimpleDirectoryReader) and network operations (e.g., fetch in custom readers) are limited to the skill's primary functional scope of data orchestration and LLM integration.
Audit Metadata