llm-app-patterns
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides educational content and production-ready patterns for LLM application development, including RAG, agent architectures, and observability.
- [INDIRECT_PROMPT_INJECTION]: The skill defines patterns that process untrusted data from external sources. Ingestion points: External documents in RAG pipelines and user queries in agent workflows (SKILL.md). Boundary markers: Prompt templates include directives such as 'Answer based ONLY on the following context' to limit the model's response scope to provided context. Capability inventory: The patterns involve LLM generation and tool usage (e.g., search, calculate). Sanitization: The provided templates use standard prompt interpolation without explicit sanitization routines.
Audit Metadata