AgentDB Learning Plugins

Pass

Audited by Gen Agent Trust Hub on Mar 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation describes the use of npx agentdb@latest to run CLI tools directly from the npm registry. These are vendor-controlled resources belonging to the author ruvnet.
  • [COMMAND_EXECUTION]: The documentation provides multiple bash commands for creating, listing, and managing learning plugins using the npx utility.
  • [PROMPT_INJECTION]: The skill implements an indirect prompt injection surface as it ingests experience data that is subsequently used for agent reasoning. Ingestion points: The adapter.insertPattern method shown in SKILL.md allows the ingestion of arbitrary experience data. Boundary markers: No explicit delimiters or safety instructions are present in the provided examples to distinguish training data from operational instructions. Capability inventory: The skill is capable of performing file system operations and executing CLI commands. Sanitization: No sanitization or validation of the experience 'pattern_data' is demonstrated in the API implementation examples.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 3, 2026, 05:44 AM