ai-vector-databases

Pass

Audited by Gen Agent Trust Hub on Jun 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is entirely focused on providing architectural advice, configuration parameters, and operational best practices for vector database systems.
  • [COMMAND_EXECUTION]: The provided code examples use standard, official SDKs for vector databases (e.g., qdrant-client, pinecone-client, pymilvus). All operations are within the scope of database management (CRUD, indexing, searching) and do not involve dangerous system calls or shell commands.
  • [CREDENTIALS_UNSAFE]: There are no hardcoded secrets, API keys, or credentials. The code examples use standard placeholders such as '...' or common prefixes followed by ellipses (e.g., 'pcsk_...').
  • [EXTERNAL_DOWNLOADS]: The skill does not contain any instructions or code to download or execute remote scripts or binaries.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 13, 2026, 09:10 AM
Security Audit — agent-trust-hub — ai-vector-databases