rag-architect

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core workflow involves ingesting and processing untrusted external data (documents) to build retrieval-augmented generation systems.
  • Ingestion points: SKILL.md (chunking implementation), references/chunking-strategies.md (various splitting strategies).
  • Boundary markers: The provided code examples do not demonstrate the use of delimiters or specific instructions to isolate or ignore embedded prompts within the ingested content.
  • Capability inventory: The skill provides templates for indexing data into multiple vector databases (Qdrant, Pinecone, Weaviate, Chroma, pgvector) and interacting with various AI services (OpenAI, Cohere, Voyage AI, Jina AI).
  • Sanitization: Explicit sanitization or content validation is not featured in the architectural reference code.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 04:40 PM
Security Audit — agent-trust-hub — rag-architect