upstash-vector-db-skills
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill documents patterns for RAG (Retrieval Augmented Generation) and semantic search which involve processing untrusted data.
- Ingestion points: The
index.upsertandindex.querymethods inSKILL.mdprocess user-provided text and external document data. - Boundary markers: The provided code examples do not demonstrate the use of delimiters or specific instructions to ignore embedded commands in the retrieved context.
- Capability inventory: The skill utilizes network capabilities to interact with the Upstash REST API via the
@upstash/vectorlibrary. - Sanitization: No explicit input validation or sanitization of the data being indexed or queried is shown in the boilerplate examples.
- [SAFE]: The skill uses official SDKs and references well-known services for vector database and embedding functionality. All external links point to established documentation and service provider domains.
Audit Metadata