AgentDB Learning Plugins

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses 'npx agentdb@latest' for plugin operations including creation, template listing, and configuration management. This command execution is part of the core functionality of the AgentDB CLI tool.
  • [EXTERNAL_DOWNLOADS]: Fetches the 'agentdb' package from the NPM registry via 'npx'. These downloads are associated with the official vendor of the skill.
  • [PROMPT_INJECTION]: The skill processes external experience data which creates an attack surface for indirect prompt injection during agent reasoning.
  • Ingestion points: External training data (experiences) enters the system via the 'insertPattern' method in 'SKILL.md'.
  • Boundary markers: The implementation lacks explicit delimiters or instructions to ignore embedded commands within the 'pattern_data'.
  • Capability inventory: The skill can execute CLI commands via 'npx' and manage local database files.
  • Sanitization: No validation or escaping of the experience data is observed before it is used to synthesize context for reasoning.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 04:33 PM