hive-mind-advanced

Pass

Audited by Gen Agent Trust Hub on Mar 1, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the claude-flow package via npx, which is the standard delivery method for the author's platform. Reference URLs in the documentation (e.g., github.com/ruvnet/claude-flow) use a non-standard delimiter ($) but correctly point to the official vendor repositories.
  • [COMMAND_EXECUTION]: Documentation provides multiple CLI examples for initializing swarms and monitoring metrics using the claude-flow utility. These commands are localized to the system's management and do not involve unauthorized shell execution.
  • [SAFE]: No evidence of prompt injection, credential exposure, or malicious persistence was detected. The use of SQLite for collective memory is standard for the described agentic architecture, and all operations align with the stated goal of multi-agent coordination.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 1, 2026, 04:33 PM