pinecone-rag
Pass
Audited by Gen Agent Trust Hub on Jun 19, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill implements a Retrieval-Augmented Generation (RAG) pipeline that is vulnerable to indirect prompt injection.
- Ingestion points: The
rag_answerfunction inSKILL.mdretrieves content from the Pinecone index (hits) and merges it into the LLM's system prompt ascontext. - Boundary markers: The code uses simple newline separation (
\n\n) to join context chunks. It lacks robust delimiters (like XML tags) or explicit instructions to the model to ignore any instructions found within the retrieved text. - Capability inventory: The skill uses the
openaiandpineconelibraries to perform search, upsert, and chat completion operations across all provided code snippets. - Sanitization: There is no evidence of sanitization, escaping, or validation of the retrieved document content before it is interpolated into the system prompt.
Audit Metadata