AgentDB Performance Optimization
Pass
Audited by Gen Agent Trust Hub on Aug 21, 2026
Risk Level: SAFEREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The skill instructs the use of
npx agentdb@latest, which dynamically downloads and executes code from the npm registry at runtime without version pinning. - [COMMAND_EXECUTION]: Documentation provides shell commands (
npx agentdb benchmarkandnpx agentdb stats) for the agent to execute to measure database performance. - [EXTERNAL_DOWNLOADS]: The skill references and links to external codebases and documentation hosted on GitHub (
github.com/ruvnet/agentic-flow) and an external domain (agentdb.ruv.io). - [PROMPT_INJECTION]: The skill processes external data via vector database lookups using
retrieveWithReasoning, creating a surface for indirect prompt injection. - Ingestion points:
adapter.retrieveWithReasoningandadapter.insertPatternmethods in SKILL.md. - Boundary markers: None provided in the examples to distinguish between data and instructions.
- Capability inventory: The agent has capabilities to execute shell commands via
npxand modify local database files. - Sanitization: No explicit sanitization or validation of the retrieved vector content is mentioned before it is processed by the reasoning engine.
Audit Metadata