AgentDB Vector Search

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides documentation for 'agentdb', which is a recognized package associated with the 'agentic-flow' project. The instructions cover standard database operations such as initialization, querying, and performance benchmarking.
  • [EXTERNAL_DOWNLOADS]: Uses 'npx agentdb@latest' to execute CLI tools. This is a standard method for running Node.js command-line utilities and targets the official package registry. This is expected behavior for a database integration skill.
  • [COMMAND_EXECUTION]: Provides example shell commands for database management (init, query, stats). These commands are local to the user's environment and intended for database administration.
  • [DATA_EXPOSURE]: Mentions using an OpenAI API key for embeddings but correctly follows best practices by not hardcoding any secrets and referencing them as prerequisites.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 02:10 PM