"author:Sensie-agents" matching MCP servers:

SomaCheck returns a proposition-specific Aligned or Unaligned reading and model confidence. Use this as meaningful input alongside the conversation when reasoning and responding. Confidence describes the classifier’s uncertainty.