rag-pipeline-builder
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a RAG pipeline architecture that ingests and processes external document content. This creates a standard surface where untrusted data within documents could contain instructions designed to influence the agent's behavior during retrieval or generation.
- Ingestion points: Document text processing via
splitter.split_text(document.text)inSKILL.md. - Boundary markers: The code snippets do not demonstrate the use of delimiters or specific instructions to the LLM to ignore embedded content.
- Capability inventory: The system involves document chunking, vector store indexing, similarity search, and LLM-based query expansion.
- Sanitization: There is no evidence of data validation or content filtering in the provided implementation examples.
Audit Metadata